Early Chinese Coinage

Yü-Ch'üan Wang
Numismatic Notes and Monographs
American Numismatic Society
New York
Worldcat Works




Open access edition funded by the National Endowment for the Humanities/Andrew W. Mellon Foundation Humanities Open Book Program.


Table of Contents



I. INTRODUCTION 1. An Historical Sketch of Ancient Chinese Numismatics
1 2. Difficulties in the Study of Ancient Chinese Coins
11 A. Decipherment of Legends
11 B. Use of Epigraphical Evidence in Dating Coins
13 C. Lack of Historical Records and Archaeological Reports
17 D. Identification of Mint Names
1. The Shang and Chou Periods 22
2. The Ch'un-ch'iu Period (770—481 B.C.) 30
3. The Chan-kuo Period (403—221 B.C.) 42
54 2. Species of the Cowries and How the Ancient Chinese Obtained Them
66 3. Imitation of Cowrie Shells
69 4. The I Pi Ch'ien (Ant Nose Money)
76 5. P'üng, the Measure of Cowries
1. The Origin of the Spade Coinage 90
2. The Date of the Coinage 100
3. The Various Types of Spades and Their Chronology 114
A. The Prototype Spade 116
B. Hollow-Handle Spade 117
C. The Old Spade 121
D. The Late Spade 123
4. The Regional Distribution of the Spades 129
5. Special Old Spades of Liang 137
V. THE KNIFE COINAGE 1. Its Origin and Date
144 2. The Early Knives and Their Distribution
157 3. The Late Knives and Their Distribution
162 A. The Sharp-pointed Knives
162 B. The Ming Knives
166 C.
The Small Knives 172
D. Dating of the Late Knives 173
E. Expansion of the Knife Coinage 176
VII. THE ROUND COINAGE 1. The Round Coins of the Knife Area
187 2. Round Coins of the Spade Area
193 3. Date of the Round Coinage
198 4. Some Historical Explanation
1. The Lieh 207
2. The Chin 211
3. The Huo 218
4. The Liang 221
APPENDIX I: Objects Wrongly Regarded as Money 237
APPENDIX II: Spades of Probable Post-Chou Origin 240
MAPS China, 5th—3rd Centuries B. C.
252 China, 8th—5th Centuries B. C.
253 Distribution of Coin Types Ca. 500—250 B. C.


In this monograph we have attempted to make a preliminary reconstruction of the monetary systems of Chinese antiquity. From the pages which follow we hope our readers will find that our attempt has been fruitful.

The reason for the choice of this topic is twofold. Firstly, practically no historical literature has been preserved which provides information concerning ancient Chinese coinages, though the coinages must have played an important part in the nation's economic, social and political development. Secondly, coins of ancient China have been found in such large numbers in recent decades that they should provide helpful data for serious works in the field of historical studies, and negligence to use them should be inexcusable.

The American Numismatic Society possesses, in our opinion, the largest single collection of Chinese coins in the world. Its officers and council have long felt the necessity of such a work as the present one as a step towards developing the scholarly potentialities of its cabinet in the advancement of numismatic and historical studies of Chinese civilization. For this reason, I was generously granted the opportunity and provided with all necessary facilities to prepare this monograph.

The collection of the ancient Chinese coins at the Museum of the American Numismatic Society consists of 4,350 specimens not including cowries, cowrie imitations and the so-called "Ant Nose Money." A part of the Museum's collection was acquired by the Society itself in the course of many years. The greater part was donated by Mrs. Eric N. Baynes, daughter of the late John Reilly, Jr., who during his life time made the collection of Chinese coins his chief interest and built up a very large and excellent cabinet. The bulk of the Reilly collection was originally that of Henry Ramsden, the famous numismatist of the early years of this century. It was on the basis of the collection at the Museum of the American Numismatic Society, occasionally supplemented by information gathered from previously published coin catalogues, that this monograph was prepared.

In preparing the monograph at the Society's Museum, I have enjoyed the complete confidence of Mr. Sydney P. Noe, Chief Curator of the Society, who relieved me from administrative duties in order to let me concentrate on my research. I benefited also from the kind encouragement and advice of Dr. George C. Miles, Curator of Islamic Coins and President of the New York City Oriental Club. Everyone of my colleagues at the Museum have extended assistance to me, especially Mr. Sawyer McA. Mosser, Secretary and Editor, and Mr. William L. Clark, Curator of Mediaeval and Modern Coins. To Mr. Mosser I must particularly express my heartfelt thanks for the revision of the manuscript. Without his help this monograph might have never reached the public.

Outside the Museum generous assistance has been received from friends and libraries. Dr. Roswell S. Britton, Professor at New York City University, was consulted on various questions. He and Dr. L. Carrington Goodrich, Professor at Columbia University, read the manuscript and made valuable suggestions which have been incorporated in the monograph. Mr. H. F. Bowker, a Fellow of the American Numismatic Society, kindly lent books from his private library, and the Eastern Asiatic Collection of Columbia University, the Chinese and Japanese Library of Harvard University, and the library of the American Geographical Society also extended to me the privilege of using their facilities. Miss Miwa Kai, Senior Assistant at the Eastern Asiatic Collection, Columbia University, frequently assisted me, and checked transliterations of Japanese personal names and titles of Japanese books.

These are only a few of the names which might be mentioned in grateful acknowledgement. The author has likewise benefited from the labor of the coin collectors, numismatists, and scholars in the general historical field of the past as well as from the academic achievements of those of the present.

If any contribution has been made to Chinese numismatics and the historical studies of ancient China in this monograph, it certainly has not been made by this writer alone.

January 1, 1950 Wang Yü-ch'üan



Though still young as a science, Chinese numismatics has had a long history. It may be said to have had its inception in the sixth century when a scholar named Liu compiled China's first coin catalogue, which was entitled Ch'ien chih (Records of Coins). According to Sun I-jang (1848—1908), the numismatist Liu was Liu Ch'ien (484—550) who is better known for his other scholarly works.1 Liu Ch'ien's catalogue seems to have been lost some time during the Sui dynasty (581—618). Our information about him and his numismatic work is derived from references to him and a few quotations from his writings by a contemporary, Ku Hsüan, compiler of the famous Ch'ien p'u (Coin catalogue).2

The text of Ku Hsüan's catalogue has also been lost. The work is recorded in the section on literature in the official history for the Sui dynasty, and passages from it are quoted in the Ch'üan chih (Records of Coins) by Hung Tsun (1120—1174).3 Judging from these quotations, Ku Hsüan's catalogue seems to have had no specimens of coins of the Chou period, and of the periods following the Chou dynasty he apparently recorded only the few coins which he had seen himself. The later the coin, the more detailed his description, which usually included the coin's design, its legend and its issuing date. His method of coin description became the pattern for later Chinese numismatists.

Ku Hsüan was followed by a few numismatists during the T'ang dynasty (618—907). Among them the most famous is Füng Yen, a scholar who is better known for his "Things Seen and Things Heard," whose work on coins is entitled Hsü ch'ien p'u, presumably a coin catalogue supplementing the one by Ku Hsüan. Although this work is lost also, enough fragments have been preserved to give us an idea of the work done in numismatics by the T'ang scholars. The first recorded discovery of Chou coins seems to have taken place in Füng Yen's time. For each coin known to have come from a find, he recorded the circumstances of its discovery and the conditions under which he was able to examine it.4 Otherwise, his statements were confined to the shape and size of the coin and the structural composition of the characters in the legend. He did not, however, venture to interpret the meaning of the legend.

Though limited in scope, the numismatists of T'ang were unpresumtuous and their reports seem to be reliable. The same cannot be said of the numismatists of the Sung dynasty (960—1279 A. D.) which followed T'ang. By their time both the Early Knife coins and the Old Spade coins of the Chou period had been unearthed. Judging from the information available, all these pieces bore legends. Puzzled by the strange archaic forms of the characters in the inscriptions and firmly believing in the legendary stories of prehistoric China which had been built up gradually since the Han time (206 B. C.—220 A. D.), the numismatists and historians of the Sung dynasty indulged in conjectural decipherment and interpretation of the Chou coin legends. They believed all these coins to be of prehistoric origin. Tung Yu (fl. 1101—1125), compiler of a coin catalogue, assigned some Chou coins as issues of the period of T'ai-hao, a legendary figure of the very early mythical history of China. Dissatisfied with Tung Yu's attribution as being too late, Lo Mi, an historian well versed in legendary history, placed a coin as early as Kuo-t'ien,5 another mythical figure who, if he ever actually lived, would have been contemporary with the Peking Man. Hung Tsun's Ch'üan chih, the earliest coin catalogue extant, was a product of this period.

However ill founded the allegations of the Sung scholars were, their opinions prevailed for some six hundred years in the field of Chinese numismatics and lasted well into the eighteenth century. Liu Shih-lu, one of the foremost numismatists of the nineteenth century, still attributed spade coins of Liang, capital of the state of Wei, of the Chan-kuo period (403—221 B. C.), to the dynasties of Yü and Hsia, both of which precede the Shang dynasty in the second millennium B. C. Another of his conjectures, likewise groundless, considered these coins as issues for use in the payment of fines. His interpretations are contained in his essay, the Yü Hsia shu-chin shih-wün,6 which later became a classic, widely read and admired by both Chinese and Western numismatists into the early years of the twentieth century. The dependence of Chinese numismatic studies on mythology did not disappear until Ts'ai Yün (1764—1824), an historian and philologist interested in numismatics, declared that the ancient copper coins preserved today were not objects of remote antiquity but were rather currencies which "flourishingly circulated during the Ch'un-Ch'iu and the Chan-kuo periods (770—221 B. C.)."7

Like Chinese classics, philology and history, Chinese numismatics witnessed an unprecedented advancement in the Manchu period (1644—1911) and produced a methodology that can be considered scientific. In the decipherment of the coin inscriptions, especially those on the coins of the Chou period, there are the famous names of Ts'ai Yün, Ma Ang and Liu Hsin-yüan.8 Of these, Ts'ai Yün is also known for his contribution to the studies of coin chronology. These scholars demonstrated that the legends on the Chou coins are not the names of T'ai-hao and Kuo-t'ien but the names of cities or towns of the Chou period. Not one of these three scholars, however, was a professional numismatist; their chief interest and contributions were historical and epigraphic.

In a narrow sense of the term, scientific numismatics was not established in China until the publication in 1864 of the famous catalogue, Ku ch'üan hui, by Li Tso-hsien.9 This catalogue contains illustrations of more than five thousand specimens accompanied by decipherment of their legends and, whenever possible, notes of their history. The author examined carefully and determined the authenticity of every specimen in his catalogue, and excluded those whose authenticity he could not verify. This careful attitude is not found in the works of any of his predecessors. It may be safely said that his is the first scientific work in Chinese numismatics. Up to the present time his catalogue is still regarded as the best and most reliable by numismatists. It constitutes the backbone of the comprehensive Ku ch'ien ta tz'ǔ-tien (Encyclopaedia of Old Coins) published in 1938.10

In spite of the progress made by scholars and numismatists in the eighteenth and nineteenth centuries, study of coins in China failed to arouse wide attention. It was pursued largely by a limited group of wealthy gentry and retired officials. These elderly gentlemen, withdrawn from the tumultuous experiences of earlier years, foundserenity of life in the companionship of antiquities. In satisfying their personal desires for large collections and in persistent search for rare specimens, they did a great service to the study of an important phase of ancient Chinese civilization in gathering the basic materials for its serious study. The only regrettable fact is that, in many cases, these men, not being trained historians and philologists, were unable to undertake satisfactorily the study of the coins. As a result, they needed the assistance of the epigraphers in deciphering coin legends. These epigraphers, in their study of the inscriptions on Shang and Chou bronzes, were led to the study of those on coins in the hope that it might help their primary work. This situation produced a somewhat bilateral development in ancient Chinese numismatics. On the one hand, the collector-numismatists studied the coin specimens but were unable to contribute substantially in deciphering the legends; on the other hand, the epigraphical scholars studied their inscriptions but neglected all other aspects of the coins. Neither group possessed the knowledge of the other, but both contributed toward the advancement of ancient Chinese numismatics. If the knowledge and the interest of both had been combined, numismatic studies in China might have advanced further.

Chinese numismatists of the nineteenth century paid little attention in their studies to the evidence of the coins themselves.11 Li Tso-hsien, the foremost Chinese numismatist of that century, did formulate a proposed systematic program for numismatic studies some ten years after his famous catalogue was published in 1864. This embraced investigation of the following: epigraphy on coins, metrology, the shapes and designs of coins, provenance, stories and anecdotes about coins and collectors, and rare specimens and specimens recorded but not seen.12 Li's proposed methodology marked a great step forward. Especially commendable is his emphasis on inquiry into the provenance of coins and their metrology. Yet, effort along these lines alone is not sufficient to enable us to exploit fully the evidence that coins can provide concerning the history of economic institutions. Study of the physical characteristics of the coins and of their inscriptions constitute the foundation of Chinese numismatics, but they alone are not enough. In order to fully understand the historical significance of the coins we must examine them against the general political, social and economic background of the period in which they circulated. Therefore, a competent numismatist must not only be an able coin examiner but an epigrapher and historian as well.13 So far, no Chinese numismatist has attained this happy combination.

In recent years, serious attempts have been made to raise the level of numismatic studies in China. In 1938, Ting Fu-pao published the coin encyclopaedia, Ku ch'ien ta tz'ǔ-tien. In this work he entered many published specimens of Chinese coins and specimens in the possession of private collectors in China. A large proportion of the illustrations are reproductions from coin rubbings. Under each specimen he has given all previously published statements and discussions concerning it, and these have been reproduced by photographic process so that they may contain no errors. It is the most comprehensive coin catalogue in Chinese.

In 1940 the Chinese Numismatic Society was established in Shanghai by a group of coin collectors and numismatists. It published a bi-monthly Ch'üan-pi, known in English as Chinese Numismatics, to promote numismatic studies. In his introductory words explaining the plans and aims of the Society in publishing the bi-monthly, Chang Chiung-po outlines a program of numismatic studies which the members of the organization are asked to follow. He first expresses his dissatisfaction with the attitude of Chinese scholars and historians who have hitherto regarded coins as "small things" unworthy of their attention. He urges rectification of this mistaken attitude and the recognition of coins as valid historical material. He suggests that "systematic researches" must be made of them in which the examination of actual specimens be made, historical records be consulted, and that coins be treated as relics of an economic and historical institution which can be studied in the light of monetary theories. Commenting on the situation in which persons writing about the history of coinage do not actually handle the coins and those who actually handle them lack the necessary scientific training to write about coinages, he states that there must be a combination of both.14

On the whole, Chang's program is well conceived, as it pointedly sets out to rectify the weaknesses in past Chinese numismatic studies. The bi-monthly appeared periodically until the end of 1945. It carried some interesting discussions and a number of articles which were written on a much higher level than ever before. But on the whole the effort of the members of the Chinese Numismatic Society fell short of the goal set by Chang Chiung-po.

However imperfect the writings of the members of the Chinese Numismatic Society may have been, their efforts in themselves are significant contributions toward advancement of numismatic studies in China. It is regrettable that the bi-monthly of the Society had to exist during the Japanese aggression in China and thus had its influence greatly limited.

In Japan, Chinese numismatics has been a favorite field with some Japanese sinologists. In that country, too, the study of old coins was generally regarded as a hobby of collectors with means and leisure, and was not accorded the attention it deserved until recently. In the last few decades Japanese archaeologists have conducted some extensive excavations in Korea, Manchuria and Jehol, and in many of the old Chinese remains they found coins of the Chou, the Han and Wang Mang Periods, with which they dated the remains. The discovery of ancient Chinese coins by these archaeological missions seems to have stimulated numismatic interest among Japanese scholars. Probably for this reason, ancient Chinese coins, not only those unearthed from the old remains but also those in the hands of collectors, suddenly acquired the dignity of archaeological objects. It was among Japanese scholars that the question of the initial date of Chinese metallic coinage was first discussed in a scholarly manner.15 These discussions did not prove very fruitful because historical scholarship was not well combined with actual knowledge of the coins, and because the historical sources were not fully understood and exhausted. However, a proper approach was made toward one of the many problems of Chinese coinage.

In 1938, the same year in which the Encyclopedia of Old Coins was published in China, the Tōa senshi (Catalogue of Eastern Asiatic Coins) by Okutaira Masahiro appeared in Japan.16 The catalogue consists of eighteen volumes of which twelve are devoted to Chinese coins. The specimens listed in it are illustrated with reproductions of well-made rubbings. Besides its typographical excellence, the work can be commended for the number of specimens of rare coins of the Chou period it includes which are not to be found in other coin catalogues. In the decipherment of a number of the controversial coin legends, Okutaira follows the suggestions of Liu Hsin-yüan and Kuo Mo-jo, noted Chinese experts in epigraphical studies of the Shang and Chou bronzes, whose decipherments have been neglected so far by Chinese numismatists. Closely following Kuo Mo-jo, Okutaira attempts to reconstruct the early history of Chinese coinage by making use of the inscriptions on the bronzes of the Shang and Chou period. We know of no Chinese numismatists who have ever systematically utilized this valuable source of information.

Another merit of the catalogue is its illustration of a few specimens of silver ingots and paper money, through which Okutaira has shown superiority over his Chinese colleagues in the realization that study of copper coins alone is not sufficient for a complete understanding of the Chinese monetary system. The work, however, contains no illustrations of coin moulds, also important material in the study of Chinese historical coinages, and a few of those which are listed are likely fabrications.

In the Western languages, books and articles on Chinese coinage and monetary history have been published in French, English and German. The names of Biot, Vissering and Lacouperie are the most prominent. While in China the effort to reconstruct a general monetary history of China was motivated by the study of the ancient coins and their inscriptions, in the West that effort seems to have been a result of literary researches. The material on which the works of both Biot and Vissering17 are based is of such a nature. There were no studies of ancient Chinese coins on an appreciable scale until Lacouperie, who in 1892 published his Catalogue of Chinese Coins. This work is the most comprehensive treatment of the earlier Chinese coins yet produced by a Western numismatist.

In preparing his work Lacouperie benefited from the intellectual heritage of the Western world. At his time, the numismatic studies of Greece and Rome had developed to maturity, and Lacouperie was able to draw upon the experience of the classical numismatists in tackling some of the problems in Chinese numismatics. This is probably the reason why he shows superiority over his Chinese contemporaries in his historical approach to Chinese coins. However, being a Westerner, Lacouperie naturally suffered from handicaps in preparing his work. He had insufficient training in the Chinese language to avoid misunderstandings of the texts of Chinese historical records.18 His knowledge of Chinese ancient history appears rather limited, and his inexperience in determining authenticity of the coins caused him to enter many spurious specimens in this catalogue.19 He does seem to have had training in Chinese philology — the "ancient script" and its evolution — which is absolutely indispensable in deciphering the coin legends or in judging the plausibility of the decipherments which have been advanced. While in most cases he follows the decipherment of Chinese scholars, he has been confident enough to make up some of his own, on which he formed his theory of the so-called "monetary unions." As L. C. Hopkins has pointed out,20 Lacouperie sometimes offers his conjectures as if they were facts, and he makes statements which actually have no foundation.21

We should not, however, underrate Lacouperie's contributions because of these shortcomings. If we read his book against the background of his time, we realize it was an admirable accomplishment. When we examine the discussions and publications which appeared after him and in which his influence can be directly or indirectly detected, we must recognize that he contributed much toward the advancement of Chinese numismatics in the Western world.

Next to Lacouperie among Western numismatists we find the name of Henry A. Ramsden. A scholar and a collector who collected to study, he promoted interest in the study of Oriental coins in general and Chinese coins in particular. "During his life," states H. F. Bowker, "he was the prime-mover in the study of the coins of his speciality, and was most probably the direct cause of the popularity which the coins of the Orient enjoyed in the United States during the last years of his life."22 In 1913 and the year following he edited the Numismatic and Philatelic Journal of Japan , the bilingual organ of the Yokohama Numismatic Society of which he was then the chairman. The journal devoted part of its pages to the study of Chinese numismatics. In that field Ramsden's interest covered a wide range from "barter money" to modern coinage, from metallic currency to paper money. A complete list of his works is found in "Ramsdeniana," in which the author praises him as "the foremost writer and most competent occidental authority in ... Far Eastern numismatics" at the time of his death in 1915.23

Since 1915 sinological studies in both Europe and North America have advanced considerably, and consequently the interest in Chinese coins has grown wider not only among collectors and museums but also among students of Chinese history. Yet, ancient Chinese numismatic studies have not progressed as might be desired.

In 1934, a group of Westerners residing in China formed the Numismatic Society of China in Shanghai, and subsequently published six bulletins, all of which deal with the modern coinage of China. In 1940, after the Chinese Numismatic Society was established, it associated itself with the latter in an attempt to combine the efforts of both organizations in the promotion of Chinese numismatic studies in China and abroad.

If, in the last thirty years, Western numismatists have not produced commendable work in the field of ancient Chinese coinage, a preparatory step has been well taken in the publication A Bibliography on Far Eastern Numismatics by A. B. Coole and A Numismatic Bibliography of the Far East by H. F. Bowker. The former bibliography is devoted to the listing of numismatic works in Chinese and Japanese, while the latter, supplementing the former, covers the literature in Western languages. Both were carefully prepared and are convenient reference works.

In the above sketch we may have appeared hypercritical in some of our remarks. It is not intended to discredit our predecessors. The scholarship of one man is bound to be limited, as are his physical energy and the scholarly achievement of his age. If, at the present, numismatists are able to see more problems and penetrate more deeply into them, this is largely owing to the advancement of historical studies in general and Chinese numismatics in particular. Without the effort of the numismatists of the past in collecting the material and preparing the preliminary studies, any new and constructive contributions would be inconceivable.

End Notes

1 Sun I-jang image, Chou-ch'ing shu-lin, 1916, VI, 19b—20b.
2 No. 64 in A Bibliography on Far Eastern Numismatics by A. B. Coole, hereafter cited as Coole.
3 Coole 112.
4 Some of these statements are quoted by Hung Tsun in his ch'üan chih (Coole 112), 1874 ed., IX, 6b, 7a, 12b.
5 "Lun pi so ch'i" (On the origin of money), Lu shih fa-hui image Ssǔ-pu pei-yao ed., I 12b—14a.
6 Coole 385.
7 P'i-t'an (Coole 342), photostat ed., II, 1b.
8 Ma Ang, Huo pu wün-tzǔ k'ao (Coole 222) and Liu Hsin-yüan, Ch'i-ku-shih chi-chin wün shu (Coole 12).
9 Coole 266.
10 Coole 240.
11 This criticism is voiced by Sun I-jang, op. cit., (see above, n. I), V, 21b.
12 "Hsü ch'üan shuo" (Supplementary Remarks to the Discussions on Coins). The Discussions on Coins or Ch'üan shuo was written by Pao K'ang, (Coole 296), 1874 ed., 17b. The article is an appendix in the Kuan-ku-kü ts'ung-kao by Pao K'ang (Coole 298).
13 Lo Chün-yü has also observed that a numismatist must have training in philology and history, and the ability to determine the authenticity of the coins. Yung-lu jih-cha (Coole 392), photostat ed., 25b.
14 See the explanatory note on the publication of the Ch'üan-pi Bi-monthly, or Chinese Numismatics as it is called in English, No. 1, 1940, pp. 2—3.
15 For discussion of the dating of ancient Chinese coinage, see pp. 100—114.
16 Coole J—162.
17 üdouard Biot, "Mümoire sur le Systüme monütaire des Chinois" Journal Asiatique, 3e sürie (1837), III, 422—465; IV, 97—141, 209—252, 441—467 and Willem Vissering, On Chinese Currency, Leiden, 1877.
18 For instance, on page xiv Lacouperie states, "Su, Prince of Tchao, grants to Tchang-y, a secret political agent of Ts'in, the privilege of issuing pu coins of the saddle-pattern." I have not been able to verify the one reference given for this statement, and presume that it relates to the story that Su Ch'in, minister of the king of the state of Chao (Tchao), persuaded the king to "give money, gifts, carriage and horse" image and send a man to secretly follow Chang I (Tchang-y) to Ch'in (Ts'in) (Shih-chi LXX, 2a). Here no grant of the privilege of issuing pu coin is involved, and Chang I was not at this time a secret political agent of Ch'in.
19 E. g., p. 16, no. 40; p. 120, no. 41; p. 121, no. 52; p. 223, no. 53; p. 224, no. 54; p.225,, no. 55; p. 226, no. 103; p. 299, no. 102; p. 298, and many others.
20 L. C. Hopkins, "On the Origin and Earlier History of the Chinese Coinage," Journal of the Royal Asiatic Society, 1895, 318.
21 E. g. his allegations that Chi-mo (Tsih Moh) was a mint of Lang-yeh (Lang-ya) which was a Western settlement on the Shantung coast (p. lxiii).
22 "Ramsdeniana," The Coin Collector's Journal, VIII (1941), 76.
23 Ibid.


A. Decipherment of Legends

Observations on the size, weight, fabric, and mint locations of coins can shed invaluable light on the early monetary history of China. The valid interpretation of any of these elements depends on the correct decipherment of the coin's legend. As we have already noted, decipherments of coin legends have in the past been made by epigraphical scholars whose main interest was ancient script and not coins. As a rule, collector-numismatists have followed the readings of these scholars, but, in cases where none had yet been made, they attempted their own. They were not trained in epigraphy, and some had not even enough knowledge of philology to determine which of several possible decipherments was the most plausible. Often a collector's guess would be colored by an eagerness to enhance the value of his coin.

The mint name image (lin, written image) in modem script) on a group of late spades has been variously read as image (lu), image (yü), image (huang), image (huang-fu), image (chia), image (huo), and image (kuan). The last reading, kuan, which was suggested by Ma Ang and publicised by Li Tso-hsien, prevailed in the numismatic world for many years. It was followed by Lacouperie and all other Western numismatists. Since neither Ma Ang nor Li Tso-hsien found Kuan as a place name in historical literature, they assumed the character to be an abbreviation for "kuan-chung," a term denoting a region which is now central Shensi. Lacouperie mistakenly called it the "capital city of Ts'in." The correct decipherment of the character, strangely enough, is said to have been made many years before the above suggestions were advanced, by Sun Hsing-yen (styled Yüan-ju, 1753—1818), a scholar in Chinese classics and philology and not a numismatist.24 Sun deciphered the character as the name of a town in the state of Chao during the Chan-kuo period. All numismatists acquainted with recent numismatic studies follow his decipherment.

Superior as opinions of epigraphical scholars are in decipherment of coin inscriptions, they are not always correct. Take the character image for instance, which appears on many Late Spades. Some numismatists read it pa-huo image or "eight huo," 25 the latter character being used here in the sense of a denominational unit. This reading is incorrect. Actually, it is a single character, not a monogram of two; and no part of the character can be construed as huo. Epigraphical scholars read it as fün image, meaning "belongs to the reign of."26 This decipherment is also wrong, for it is not suggested on the basis of the character's structural identity with fün, but on its resemblance to it. The correct reading is pan image (half), a denominational term (with the name of the unit understood) of the coinage of the Chou period. When these coins were in circulation there were only two denominations, a full unit and a half, the unit being chin image. The weight of the coin bearing the legend pan (half) is just half that of the full unit piece.27 Had the epigraphers gathered and weighed specimens of both sizes they would not have escaped this conclusion. The same holds true for the deonominational term ling image which is found on the larger spades of the state of Ch'in (Late Spade IV) and which has been improperly deciphered.28 By weighing spade coins of Liang of various sizes, Kuo Mo-jo, an able contemporary epigrapher, ascertained the correct reading of the eight-character legend which they bore.29 It was only after an investigation of the provenance of a group of late square foot spades that Okutaira accepted Kuo Mo-jo's suggestion that their legend reads "Hsiang-p'ing," a mint located in present day southern Manchuria.30 These examples show how important it is to combine epigraphical and philological with numismatic evidence. The use of one of these types of evidence to the exclusion of the others constitutes one of the chief obstacles encountered in numismatic works of the past as well as the present. It is this situation which is responsible for many unacceptable decipherments, which will have to be reconsidered or discarded in the present study.

End Notes
24 Quoted by Hsü Yüan-k'ai (Ku ch'ien ta-tz'ǔ-tien, XIII, 499a).
25 Such as Li Tso-hsien, op. cit. (see above, n. 9), yüan III, 1b.
26 Such as Liu Hsin-yüan, op. cit. (see above, n. 8), XIX, 7b ff. and 23b.

B. Use of Epigraphical Evidence in Dating Coins

Some numismatists may cherish the idea that a comparison of the epigraphy of the coins with that of the bronze vessels of the Shang and Chou periods should furnish criteria for dating the coins. This task, however, is not as easy as it appears at first glance.

There are two aspects of ancient Chinese epigraphy: (a) the structural form of the character and (b) the style or manner of executing the character. So far as the structural form of the characters is concerned two changes have taken place which have bearing on the epigraphical chronology of Chinese script. In the course of time, many characters have undergone simplification while many others have become more complicated through acquisition of signifies (i. e., radicals signifying "water," "walled city," etc.). In other words, with some characters, the more complicated their form the older they are; with others, the simpler their form, the older they are. On coins of the Chou period we sometimes find simpler forms of characters of the second group in inscriptions of pieces of a considerably late date, when the complicated form of the character had become the norm. Because of this circumstance, dependence on coin epigraphy alone will fail to determine correctly the date of a coin.

The style of Shang bronze inscriptions is characterized by a peculiar execution of the strokes. Each end of the stroke is usually very thin while the central part is broad and thick. This style of script has been called the K'o-tou, or tadpole, script. However, the tadpole script is found also to be the dominant style in inscriptions of the early years of the Chou period. Without considering its content along with other factors, even an expert epigrapher cannot determine to which period an inscription in the script belongs.

The period of the Chou dynasty is, for convenience sake, usually divided into three smaller periods: The Western (Early) Chou period covering almost three hundred years from 1122 B. C. (traditional date) or 1027 B. C. to 771 B. C.; the Ch'un-ch'iu period from 770 to 481 B. C.; the Chan-kuo period ending in 221 B. C. During the first period bronzes were almost entirely made by or for kings and ministers of the royal court; during the second and third periods they were made practically only by rulers and nobles of the various feudal states. The style of the inscriptions of these three periods were roughly the so-called Ta-chuan (great "seal" character), the Chou-wün (slightly simplified Ta-chuan), and the Hsiao-chuan (small "seal" character). Whether the Ta-chuan can be regarded as also the style used in the various feudal states during the first period and whether the other styles found in the various feudal states can be regarded as also the style in the royal domain of Chou during the second and the third periods cannot be said with absolute certainty, especially when we realize that, though so simply stated above, the styles of script during the second and the third periods of Chou present strong local divergence. All these factors complicate any attempt to use comparative study of coin inscriptions and bronze vessel inscriptions in determining dates of coins.

Even if we disregard these complications, dating of coins by comparative study of inscriptions on bronzes is made impossible by another circumstance. Coins and bronzes have different epigraphical styles which result from the difference in their purpose and in the techniques of inscribing them. On bronzes, inscriptions were cast on the vessels, which were made in honor of the maker's forefathers or other relatives, to commemorate a victory in war or a royal or princely grant, to glorify his enfeudation as a prince or his appointment to an office, to record an important event or a settlement of a dispute. The personages involved are always kings, princes and upper class nobles. Some of the vessels bearing the inscriptions commemorating enfeudations were kept in the ancestral temples by princes and venerated as symbols of the existence of their states. They thus had a monumental character and their inscriptions were accordingly rendered in a conventional and elegant manner.

The coins were not personal treasures of kings, princes or nobles; they were made to be used as media of exchange in a society, the overwhelming majority of which was illiterate. Inscriptions on the coins are merely marks indicating the mint's name, sometimes the serial number of minting, and occasionally also the denomination. To coins of full intrinsic quality, whose value depends largely on their alloy and weight, these marks are not essential; accuracy and elegance in style of character are matters of secondary importance. That is probably the reason why the inscriptions on the coins are generally crudely rendered, while those on bronze vessels are usually models of calligraphy. A comparative stylistic study between crude script and highly developed calligraphy is hardly possible.

Furthermore, whereas the inscriptions on the bronzes were, as generally acknowledged, written by persons with training in calligraphy, those on the coins were left to artisans at the mint, who did not always follow the conventional style and contracted the structure of the characters to the extreme. Technically, inscriptions on bronze vessels were cast from a mould which was made from a model, on which both the designs and the inscription were carved out to the desired fineness. The inscription on the coin was, on the other hand, cast from a mould which was not made after a model and on which the inscription was carved directly and in reverse. As a result, the strokes of the characters on coins were generally in contracted straight lines, for these were much easier to make than curved lines. The straightening of lines and consequently the contraction of the structure of the characters further reduces the possibility of a satisfactory comparative stylistic study of the coin inscriptions with those on bronze vessels.

Can we detect an evolution of style within coin inscriptions themselves with which we may find out the order of appearance of the coins? This question also brings complications. The fact is that the coins were cast by local mints many of which undoubtedly belonged to princes and minor nobles and even wealthy private individuals. Under these circumstances local character and individual inclinations could not but exert their influence. T'ang Lan, a contemporary epigrapher who has specialized in the study of inscriptions on oracle bones and bronzes of the Shang and Chou dynasties, holds the opinion that during the Chou period the style of script in common use was much more simple and irregular than the official script and that it had influenced the official writings towards the end of the period.31 Local variances render the study of the evolution of the style of coin inscriptions difficult if not altogether impossible. The curious thing is that we find some of the most archaic forms of characters on comparatively late coins cast in the border regions of Chou China, where, sociologically speaking, as in modern colonies, conservatism was usually stronger than the central area.

However, this does not mean that epigraphical studies have no bearing at all on the determination of coin chronology. On the whole, we may say they do, but only in a general way. They can be applied only in the cases in which stylistic distinctions can be positively established and this is possible only with either the very old or the very late coins. The older the coins are, the closer is the style of their inscriptions to the Ta-chuan (great seal character), and the later the coins, the closer to the Hsiao-chuan (small seal character). The former is identified with the epigraphical style of the inscriptions of the vessels of the Western Chou period, and the latter is the style officially adopted and made universal in 221 B. C. In structure, the Hsiao-chuan is much simpler.

End Notes
27 Okutaira Masahiro has also found this correct decipherment, op. cit. (see above, n. 16), III, 15b.
28 Li Tso-hsien reads it as tsai image meaning "the official of the town of," in other words, the official of the mint whose name appears on the obverse of the coin. Op. cit. (see above, n. 9), yüan IX, 4b. For a correct explanation of the denominational term see p. ooo.
29 Kuo Mo-jo image Liang Chou chin-wün tz'ǔ ta-hsi k'ao-shih, Tokyo, 1935, 13b.
30 Ibid, IV, 15b.
31 T'ang Lan image Ku wün-tzǔ-hsüeh tao-lun, Peiping, 1935, I, 51a.

C. Lack of Historical Records and Archaeological Reports

Another difficulty in studying ancient Chinese coins is the lack of literary records. Excepting the simple terms of pei (cowrie), tao (knife coin) and pu (spade coin), other data regarding the ancient Chinese monetary system is not found in historical literature. The widely believed traditional story about the casting of the "big coins" by King Ching of Chou in 524 B. C. is embodied with unreliable elements (i. e., the remarks made by Shan-mu-kung). The anecdote that King Chuang of Ch'u (613—519 B. C.) aroused resentment among his people by replacing "light" coins with "big" coins, which also has been regarded as factual, must be accepted with reserve. The story recorded in the Kuan-tzǔ and the "present edition" of the Bamboo Annals that Ch'üng-T'ang, founder of the Shang dynasty, cast metallic money is pure folklore. Even if these stories were reliable, they still contain no essential information on ancient Chinese coinage. The only reliable material in our possession with regard to monetary systems in Chinese antiquity are the coins themselves, but some problems presented by them are hard to resolve because of the lack of historical records.

The major problem that suffers from lack of historical records concerns the date of the origin of coinage in China. The difficulty in dating the earliest coins would be also considerably less if the conditions were known under which the coins had been discovered.

What knowledge we have about discoveries of coins is scanty. In the scientific excavations of the Academia Sinica at early Chou and pre-Chou sites in North China only cowries were found.32 The excavations at Ch'üng-tzǔ-yai in eastern Shantung produced only a broken handle from an early knife coin.33 Ming knives were unearthed among old remains in I county of Hopeh Province by an expedition led by Ma Hüng in 1920.34 These knives, however, were very late in date. Mr. Kuo Pao-chün of the Academia Sinica has reported to the author that a number of pointed-foot hollow-handle spades were discovered in Chou tombs of Chün County, Honan. During the last fifty years Japanese archaeologists have excavated both spade and knife coins of the late Chou period in Jehol, Manchuria and Korea.35 Except for those reported by Kuo Pao-chün and the handle of the knife coin, there are no reports by excavators of the discovery of early coins.

Coins of the earlier periods have been found casually, for the most part by farmers while tilling their fields. From the farmers they passed to collectors, either directly or through coin dealers. Thus, by the time they reach the hands of collectors they are isolated objects, completely dissociated with the site of discovery and the other objects with which they had originally been deposited. This condition would not obtain, had their discovery been made under the trained observation of archaeologists. The situation becomes the more regrettable when we realize that tens of thousands of Chou coins have been casually retrieved and, so to speak, lost again.

Among the late spade coins there is a group, which, as forerunners of the pan-liang (half liang) round coins of imperial Ch'in, are important for the reconstruction of early Chinese coinage. The group is of the round-footed type with three holes (Late Spade IV). Their monetary unit is the liang, as is specified on the reverse legend, and so far one liang and half liang (i. e., 12 chu) pieces have been found. Fifteen specimens from nine different unidentified mints are known.36 If the places of their discovery and the conditions of their finding were known, it might be possible to locate their mints with some certainty.

End Notes
32 See An-yang fa-chueh pao-kao image T'ien-yeh k'ao-ku pao-kao image and Chung-kuo k'ao-ku hsüeh-pao image
33 Ch'üng-tzǔ-yai image Nanking, 1934, 89 and Plate LII, no. 9.
34 Fu Chün-lun, "Yen hsia-tu fa-chüeh pao-kao," Kuo-hsüeh chi-k'an image III (1932), 180.
35 See the volumes of the Archaeologia Orientalis image a series published by the Tōa Kōkogaku Kwai from 1929 on. The results of these discoveries and those made in Korea have been summarized by Fujita Ryōsoku image (1892 —) in his "Chōsen hakken no meitō-sen to sono iseki," Keijo Teikoku Daigaku Bungaku Kwai ronsan image No. 7, Shigaku ronsan, image 1 —88, 1939.

D. Identification of Mint Names

The fourth major difficulty in the study of the Chou coins lies in the identification of the mint names. This difficulty arises not from the lack of historical information, but from, so to speak, the mass of it. With the exception of a few early spades, the coins of the Chou period, be they knives or spades, usually bear a legend or legends on both their obverse and reverse. Part of, or the entire, legend on the obverse is the name of the mint which cast the coin. By locating these mints a reconstruction of the distribution of the coin types can be achieved and the problem of the right of coinage of the minor feudatories can be investigated. But during the Chou period, towns in different states, and sometimes in the same state, may have the same name.37 For instance, we find "An-yang" as a mint name in the legend on one group of the early knife coins. In literary sources we find three towns of this name. There are four towns with the name "Chung-tu," which is found on a group of square-foot late spade pieces (Late Spade II). For "P'ing-yang," which appears as the mint-name of some square-foot late spades, there are five towns. Seven are found for "Hsin-ch'üng" (meaning "new city"), a mint name on a group of point-footed late spades (Late Spade I). Such examples would make a long and tedious list.

To identify the mints with towns recorded in historical literature is not easy, and there is no literary information available which will help. In solving the question we must depend mostly on our knowledge of the coins themselves. With some degree of certainty we may presume that coins within a general given area will have similar shapes and designs, that they will have the same denominational system. Therefore, by studying these aspects of the coinage of towns neighboring the possible mint we may be able to identify and locate it. Thus, we reach the conclusion that the mint An-yang of early knife coins must be the one located in present southeastern Shantung for the reason that during the Chou period towns outside of that province did not use the early type of knife coin. The An-yang of the round-footed spades with three holes (Late Spade IV) is most likely the one located in present northern Honan which was captured by Ch'in in 257 B. C., for this type of spade coins is probably of Ch'in origin. We can approach the tentative solution of the locations of Chung-tu and P'ing-yang by the same method.

Not all of the mint names which appear on the coins can be found in the literary sources. As a matter of fact, many cannot be. In such cases, to reduce our difficulties in locating the mints to a possible minimum, we may resort to the locations of the mountains, rivers, and other landmarks after which early settlements in China were often named. For example, "Chi-yin," a mint name on some of the round coins of Chou, is not found in the literature of this period. The meaning of the place name is "on the yin side of the Chi." This means that the town in question was located on the yin side of either a mountain or a river by the name of Chi. In Chinese antiquity there was no mountain of this name, but one of the four most important rivers was so designated. It traversed the western part of present Shantung. The yin side of a river is its southern side, and therefore, the town in question must have been located on the southern side of the Chi River. Since this town was named Chi-yin, there may have been also a town named Chi-yang (meaning "on the northern side of the Chi River"). Luckily, a town by this name did exist during the Chou time and is recorded in the contemporary literature.38 It was located northeast of modern Lan-füng in eastern Honan (approx. 115E and 35N). Judging from their names, the towns of Chi-yin and Chi-yang may have been opposite to each other, or at least they must have been located in the same neighborhood. This assumption is confirmed by the location of a city also named Chi-yin in the third century B. C. It was the capital of a Han province of the same name. The city was located about a mile northeast of modern Ting-t'ao in southeastern Shantung, and about thirty miles northeast of old Chi-yang. In all probability, the Chi-yin of Han may have been the Chi-yin of Chou. Thus, by resorting to landmarks we can locate Chi-yin and other mints whose names, though not to be found in ancient literature, have a geographical origin.

In connection with the identification of mint names mention should be made of the practice among Chinese numismatists of regarding some single character legends on ancient coins as abbreviations for two character mint names. For instance, mu has been considered an abbreviation for Mu-mün, yang for Kao-yang, li for Kung-li, kung for San-kung, etc. It is true that a mint name can be abbreviated for lack of space. The abbreviation of "Chin-yang" as "Chin" on a Small Knife is an example. It, however, can be proved, while those mentioned above cannot. Unless a claim of abbreviation can be proved, it must not be accepted without reserve.

End Notes
36 For the illustrations of these specimens see Okutaira, op. cit. (see above, n. 16), IV, 71b —74a and Ku ch'ien ta-tz'ǔ-tien, VII, 406b, no. 1226.
37 For a general idea of the situation of confusion see Ku Tung-kao image (1679—1759), Ch'un-ch'iu ta-shih nien-piao, 1752, VI, Part 2, 25a —34a for the Ch'un-ch'iu period. For the Chan-kuo period see Ku Kuan-kuang image (1799 —1862), Ch'i-kuo ti-li k'ao.
38 See Chu-shu chi-nien image (Wang Kuo-wei image (1877—1927), Ku-pün chu-shu chi-nien chi-chiao in Wang-chung-ch'io-kung i-shu, second series, p. 16b.). The town belonged to the state of Liang (Wei image) and was walled in 341 B.C.



A brief survey of the development of commerce in ancient China seems necessary to provide a general background for our discussion of the origin and evolution of Chinese coinage, for there are no works, either in Chinese or other languages, which can be recommended for reference on this topic.

According to an old myth, markets were established in prehistoric China by a legendary hero, Shün-nung. It is idle speculation to consider the possibility of commerce at such a remote time, since its significance in the general economy could not have been any greater than that of barter trade among present day primitive peoples. It will be more profitable to study the economic conditions under the Shang dynasty, for which period there is a fair amount of archaeological devience.

Some scholars have asserted that the Shang economy was based either on cattle-breeding or on a combination of cattle-breeding and rudimentary agriculture. Others would have us believe that the Shang people lived in an even more primitive state. However, study of oracle bone inscriptions has proved rather the contrary.1 By conquest and colonization the Shang had built up a large empire. During the last two hundred and fifty years of their history, which ends in 1122 B. C. according to the traditional chronology, their territory extended to the sea in the east, to central Shensi in the west, to southern Hopeh in the north, and to the banks of the Huai River in the south.2 The people of this empire led a life which was predominantly agricultural.3 Recently a Chinese scholar suggested the possibility that ploughs pulled by oxen were used to till the fields and that bronze ploughshares were known.4 These ideas are challenging, even though there is no positive evidence to prove them.

In his campaign against the "Kuei-fang" state, northwest of the Shang kingdom in modern Shansi province, King Wu-ting called to arms 23,000 of his subjects in a period of three months.5 In one of his expeditions against the Ch'iang people in the west, he conscripted 10,000 on a single day together with 3000 more from a vassal state.6 Conscriptions on such a scale could not have been possible unless there had been a fairly large population, and the existence of a large population presupposes a developed economic production, at least in agriculture.

In handicrafts the Shang people achieved exceptional skill and great delicacy of aesthetic taste, as is widely demonstrated in their beautiful bronzes. Their skill in casting finds no match in later periods of Chinese history.7 The excellence of their workmanship has caused Creel to claim that it can barely be surpassed by today's metal worker with modern science and technology at his command,8 and that it has seldom been attained "anywhere in human history."9 Creel's opinions may be accented with enthusiasm, but they testify to the high quality of the products of Shang craftsmen. Such a degree of technical accomplishment could have been attained only through specialization, which in turn, could result only from division of labor. The presence of these two factors, specialization and division of labor, precludes the possibility that each family unit was economically self-sufficient. It obviously points to the existence of an economy based, at least partly, on exchange.

The economy of the state as a whole showed a similar dependence on products of other regions. The basic metals, copper and tin, which were used in the manufacture of weapons, sacrificial vessels, and many other utensils, were not to be found within the boundaries of the kingdom. They had to be obtained from the south, in and beyond the Yangtze Valley.10 The same was true of gold, silver and cowrie shells,11 which last were used both for ornaments and as media of exchange. Their red pigment, known as cinnabar, came from Shu (modern western Szechuan) and their jade was imported from western regions far beyond the Shang borders.12 Most of the tortoise shells, which were highly prized and served for divination, were of non-local origin; some of the larger specimens may have come from as far away as Malaya.13

It is very likely that the Shang kings obtained a good part of these imported products as tribute from southern states subject to them, or as booty. As the amount obtainable from such sources could not have satisfied the demands of both the court and the people, a certain proportion of the products must have been acquired through exchange.

The mutual dependence of the various regions within the kingdom was even more evident.14 Take salt for example. There were probably only two sources for this commodity. One was lake salt from present southwestern Shansi and the other sea salt from the east coast. At the present time there are no other areas within the limits of the Shang dominion which produce salt in any quantity, and it is probable that there was none in ancient times. In addition to salt there were no doubt other necessities of daily life which were specialties of particular regions also.

Due to lack of archaeological or literary evidence we are ignorant of the extent of exchange in this early period. We can, though, conclude with confidence that commerce had reached an active stage. Since Han times the word shang has been used to designate "trade." The explanation of Han scholars that shang (i. e., tradesmen) refers to traders who travel long distances is a rationalized interpretation of the word rather than an exposition of its original meaning. Recent research has shown that the character "shang" was used in ancient China only to denote the dynasty, the people, or their capital. Hence, Hsü Chung-shu identifies the term shang-jün in the sense of "tradesmen" with shang-jün meaning "people of Shang."15 In his opinion, after their conquest by the Chou, the Shang people found themselves relegated to an inferior position, which circumstance forced many of them to take up trading, an occupation deemed degrading by the upper classes. Hsü draws a parallel between the Shang people and the Jews, both being peoples forced by circumstance into trade as their special profession. While he may be correct in this conjecture, it is equally possible and even more plausible that the identification by the Chou of Shang natives with tradesmen took place a few centuries earlier when the Shang, at the peak of their prosperity, came to the more backward Chou to exchange their own products for those of the tribes of the west. Assuming this as true, it would be only natural for the Chou to identify trading as an outstanding characteristic of the Shang.16 If this interpretation is plausible, we may venture that a group of professional merchants existed in the Shang state whose business extended well beyond their own borders.

That the Shang people had engaged in widespread trade can be inferred from a statement in "Chiu kao," a decree issued by King Wu,17 the founder of the Chou dynasty, ordering the vanquished Shang of the "Mei State" to cease their overindulgence in wine and to devote themselves to farming and trade. The decree says, "You should, working hard, take your carriages and oxen and pursue trade over long distances so that you can filially nurture your fathers and mothers."18

A statement of Confucius indicates that the Chou people, who possessed a cruder culture, absorbed the Shang civilization after its conquest.19 This is corroborated by both archaeological and literary evidence. In the economic sphere, likewise, they must have inherited the pattern of the people they conquered. Unfortunately, with the exception of a few inscriptions on early Chou bronzes, we have practically no information on the economic life of the first two hundred years after the change of dynasties.

For the later Chou period, reference to trade is made in the Ode of Chan-yang which has been preserved in the Shih ching or Book of Odes. 20 This ode expresses grievances of the people against misconduct of government, interference in politics by women, and pursuit of trade by the nobility. A part of it runs:

Such things as trade yielding three times (its capital), A superior man should have knowledge of. A woman has nothing to do with public affairs, Yet she leaves her silkworms and weaving.21

This is a description of behavior contrary to the accepted norm. It was considered degrading for a nobleman to engage in trade, but obviously the temptation of three hundred per cent profit was hard to resist.

The woman referred to in the ode is said to have been Pao Ssǔ, first a court lady of King Yu (781—771 B. C.) who was made his queen when he ascended the throne. Through her influence, he is reported to have misruled his people and to have invited the Jung invasion which almost ended the Chou. If this identification is correct, the ode must have originated in the eighth century B. C.

End Notes

* A few of the works quoted in this and other sections have been translated into English and French. The Shang-shu has been translated by Legge and Karlgren under the titles of Shoo King (Chinese Classics III) and "Glosses of the Book of Documents" (BMFEA No. 20) respectively. The Shih ching (Mao Shih) has been translated by Legge, Waley and Karlgren under the titles of She King (Chinese Classics IV), Book of Songs (incomplete), and "Book of Odes" (BMFEA, Nos. 16 and 17). The Lun-yü, the Müng-tzǔ, the Tso chuan, and the Li chi have been translated by Legge under the titles Confucian Annalects, Works of Mencius, The Ch'un ts'eu with the Tso chuen (Chinese Classics I, II, V), and Li Kü (Sacred Book of China , IV and V). The first forty-seven chapters of the Shih-chi have been translated by Chavannes under the title of Les Mümoires historiques des Se-ma Ts'ien. Portions of Chapter XXX and CXXIX of this work and parts of Han shu XXIV have been translated by R. C. Blue in Harvard Journal of Asiatic Studies II. Part of the Hsün-tzǔ has been translated by Dubs under the title of Works of Hsüntze, and part of the Mo-tzǔ has been translated by Yi-pao Mei under the title of The Ethical and Political Works of Motze. The Yen t'ieh lun has been translated by Gale under the title of The Discourses on Salt and Iron.
1 This point is also well illustrated by the change of opinion in the works of Kuo Mo-jo image one of the leading authorities on the institutional history of the Shang dynasty. In his Chung-kuo ku-tai shü-hui yen-chiu, published in 1930, the author categorically declares, "There is no doubt that the Yin (Shang) dynasty was a period during which cattle-breeding was most flourishing" (p. 245), and "Although agriculture had been discovered, it was not fully developed" (p. 254). But in his Shih p'i-p'an shu, published in 1945, he rejects his former opinion and declares that during the Shang dynasty "agriculture had actually become predominant" (p. 13).
2 The boundaries of the Shang kingdom can be traced from the locations of its vassal states and of the countries it attacked and conquered. The names of these states and countries are found in inscriptions on oracle bones discovered at Yin-hsü, the site of the last Shang capital. See Tung Tso-pin image Yin li p'u (calendar of Yin), 1945, Part II, IX, 37b —40b and 61a —63a; Hu Hou-hsüan image "Pu-tz'ǔ chung so-chien chih Yin-tai nung-yeh" (Agriculture of the Yin dynasty as seen in the inscriptions on the oracle bones), Chia-ku-hsüeh Shang-shih lun-ts'ung, Second Series, 1945, 31a —47a; and Ch'ün Müng-chia image "Shang-tai ti-li hsiao-chi" (A note on the geography of the Shang dynasty), Yü-kung image (Chinese historical geography) VII (1937), Nos. 6 —7, 101 —108. In 1935 Prof. Fu Ssǔ-nien image published his essay, "I Hsia tung hsi shuo" (Ch'ing-chu Ts'ai Yüan-p'ei hsien-shüng liu-shih-wu-sui lun-wün-chi, 1093 —1134), in which he says that at the peak of its power the eastern boundary of the Shang empire extended to the "east of the sea" (meaning part of Korea) in the east, to the "south of the Ch'i mountains" in central Shensi in the west, and to the valley of the Huai river in the southeast. In other words, the territory of Shang covered present Hopei, Shantung, Northern Kiangsu, Northern Anhui, Honan, Southern Shensi, and Eastern Shensi. His study is based entirely on literary sources, and, except for his claim on the northeastern boundary of Shang, which has not been otherwise proved, is confirmed by studies made of oracle bone inscriptions.
3 Hu Hou-hsüan, ibid.
4 Hu Hou-hsüan, op. cit. 80b —81a.
5 Tung Tso-pin, op. cit. Part II, IX, 38a. This was compiled by Prof. Tung according to his reconstructed Shang (Yin) calendar. Its final validity depends on that of his calendar.
6 Tung Tso-pin, op. cit. 39a and 40b. The bone inscription quoted by Tung Tso-pin is no. 310 in The Couling-Chalfant Collection by F. H. Chalfant, Shanghai, 1935.
7 See T'ang Lan image "Chung-kuo ku-tai mei-shu yü t'ung-ch'i" (Art and the Bronzes of Ancient China), Chung-kuo i-shu lun-ts'ung (Essays on Chinese Art), ed. by T'üng Ku image Ch'ang-sha, 1938, 111 —113; Hsü Chung-shu image "Kuan-yü t'ung-ch'i chih i-shu" (On the Art of the Bronzes), op. cit. 125 —137; and Hu Hou-hsüan, "Chung-yang-yen-chiu-yüan Yin-hsü ch'u-t'u chan-p'in tsan-kuan chi" (A Note on the Exhibit of the Objects Recovered at the Yin Ruins by Academia Sinica), op. cit. 157 —167.
8 Creel, Birth of China , New York City, 1937, 112.
9 Creel, op. cit. 124. In his Studies in Early Chinese Culture (1937) 233 Creel remarks, "Chinese bronze vessels are equal to the finest objects of the sort ever produced anywhere by man. Shang bronze vessels, as a group, are probably the finest of Chinese bronzes. Among the Shang bronzes excavated by the National Research Institute in 1934 and 1935 are complicated vessels which show a genius of design and a complete mastery of technique such as to take the breath of a hardened connoisseur."
10 The "Yü-kung" in the Shang-shu states that Yang Chou (in the Yangtze Valley) produced "three kinds of metal," said to be gold, silver and copper. Ssǔ-ma Ch'ien image states that gold, tin and lead were produced in Chiangnan (south of the Yangtze River) and copper was produced in the region of Wu (lower stream of the river). Shih-chi, Po-na-pün ed. CXXIX, 1b and 11a. W. Yetts believes that copper was once produced in the interior of ancient China. Local tradition says that long ago copper and tin as well as other metals were mined some forty li northwest of An-yang city (in Honan) from the T'ung Shan or Copper Hills. Two other place names, T'ung Shan Chün and Nan T'ung-yeh (Southern Copper Foundry), testify to the tradition (An-yang: A Retrospect, The China Society, London, 1942, 25. Prof. L. C. Goodrich kindly furnished this information).
11 For gold and silver see the above note. The Kuan-tzǔ states that gold came from the valleys of the Ju and the Han rivers. Ssǔ-pu pei-yao ed., XXIII, 3a. The Han River was a branch of the Yangtze. The cowrie shells which are discovered in large numbers at Yin-hsü are Cypraea moneta and C. annulus. They may have come from the Ryukyus, the Malay Peninsula, the T'zord Banks, or as far as the coast along the Indian Ocean. See pp. 55 — 56 and 66 — 69 for more details.
12 In the excavation of the remains at Yin-hsü in An-yang there have been found a number of inscribed oracle bones painted with red and black pigments. The red pigment has been identified as raw cinnabar through chemical analysis by Professors Beneditti-Pichler and Gettens (see Dr. R. S. Britton, Fifty Shang Inscriptions, 1940, 7). Throughout ancient China cinnabar was the chief material used for red paint, as witnessed by the statement of Li Ssǔ (d. 208 B. C.) in his memorial to the king of Ch'in in 237 B. C. (Shih-chi, LXXXVII, 4b). At a somewhat later date it became associated with Taoist magic. Cinnabar is produced in both Szechuan and Hunan, but Szechuan, anciently known as Shu, was the producing district in Chinese antiquity.
13 Tortoise shells which the Shang people used for divination and which they used in large quantities were not produced within the Shang territory; they came from the Yangtze valley and the farther south (See Hu Hou-hsüan, "Yin-tai pu-kuei chih lai-yüan" or "The Origin of the Divination Tortoise Shells of the Yin Dynasty," Chia-ku-hsüeh Shang-shih lun-ts'ung or "Essays on the History of the Yin (Shang) Dynasty based on the Study of the Oracle Bones," first series, 1944, Vol. 4, 1 ff.) Prof. Tung Tso-pin quotes Wu Hsien-wün to the effect that the large tortoise shell of the Wu-ting period discovered at the old remains of the Shang capital resembles the species found today in the Malay Peninsula ("Tsai-t'an Yin-tai ch'i-hou," or "Again on the Weather during the Yin dynasty," reprint from the Studia Serica, p. 16, and Lien-shüng Yang, "Ten Examples of Early Tortoise-shell Inscriptions," Harvard Jour. of Asiatic Studies, XI, 1948, 122.)
14 A general picture of the local products in ancient China can be gathered from the statements in the "Yü-kung," a section in the present text of the Shang shu, and from those in the "Huo-ch'ih chuan" (CXXIX) in the Shih-chi. The former was written during the Chan-kuo period and the latter was written about 100 B. C. Sun Yüan-chüng image has selected various items from the above mentioned works and compiled a table showing the distribution of raw materials and the products of industry in different regions of ancient China, Yü-kung, I (1934), No. 3, 26 —38. The local products recorded in the Shih-chi are quoted in the text below. Those recorded in the "Yü-kung" are roughly salt, lacquer, embroidery, silk, the ch'ih linen from the east; lumber from tall trees, gold, silver, copper, ivory, hides, feathers, big tortoise shells, pearls, and oranges from the south; iron and silver from the southwest; and various jades from the west.
15 Hsü Chung-shu, "Ts'ung ku-shu-chung t'ui-ts'ü chih Yin Chou min-tsu" (A tentative study of the peoples of the Yin and the Chou based on the ancient literature), Kuo-hsüeh Lun-ts'ung image I (1927), 109 —113. The literary datum on which Hsü Chung-shu bases his identification of merchants with the Shang (Yin) people is in Tso chuan, Ssǔ-pu pei-yao ed., XLVII, 8b —10b. In 1937 Ojima Sukema image a Japanese scholar, published an article on the origin of the name, shang-jün. Using the same document, he identifies the first Chinese merchants with the conquered Shang people. This article forms part of his recent book entitled Kodai Shina kenkyū, Tokyo, 1944, 138—154.
16 Kuo Mo-jo attributes the origin of the twofold meanings of the term shang-jün as "Shang people" and "tradesmen" to the possibility that the Shang people may have been the first traders (Shih p'i-p'an shu, 16.)
17 Some other scholars regard King Ch'üng, son of King Wu, as the one who issued the "Chiu kao." Which opinion is correct is not material, for King Ch'üng ascended to the throne in the seventh year after his father conquered the Shang nation, a date which is very close to the Shang period. The decree is contained in the Shang-shu, known in the West as the Book of History or the Book of Documents.
18 Shang-shu, Ssǔ-pu pei-yao ed., VIII, 6b.
19 Confucius says that the Chou people followed the li (institutions) of the Yin (Shang) dynasty. Lun-yü, Ssǔ-pu pei-yao ed., ii, 6a.
20 The Shih ching contains many odes originating in the 11th to 6th centuries B. C. Except for a limited number, the odes are songs of the people, and as such they reflect actual living conditions. (Karlgren takes exception to this interpretation. He believes that the odes are too elaborate to be products of farmers. See his "Glosses on the Kuo Feng Odes," BMFEA 14 (1942), 75. Prof. Goodrich kindly furnished this information).
21 Shih ching, Ssǔ-pu pei-yao ed., XVIII, 24b. Legge's translation of these lines (Chinese Classics, 1871, iv, Part 2, 561 —2) reads: As if in the three times cent, per cent, of traffic, A superior man should have any knowledge of it; So a women who has nothing to do with public affairs, Leaves her silk worms and weaving. Karlgren's translation ("Book of Odes" BMFEA, No. 17, 1945, 86.) reads:They are like those who sell at a triple profit; the noble man knows this, and (therefore) the women have no public service, they have to (rest) abide by their silk worm work and their weaving. Legge's version is closer to the meaning of the Chinese text.

2. THE CH'UN-CH'IU PERIOD (770—481 B. C.)

In the early years of the eighth century B. C., Duke Huan (806—771 B. C.) of the State of Chüng entered into a sworn agreement with the merchants in his territory. Such an event presupposes that a flourishing trade had become important enough to elevate greatly the social position of the merchant class. An account of this agreement was made by Chüng Tzǔ-ch'an, a member of the Chüng ruling family and the most famous statesman of Chüng, to a high minister from the state of Chin in 526 B. C.22 According to Tzǔ-ch'an, when Duke Huan moved from the west to the east and established his state in what is now central Honan,23 he concluded an agreement with the merchants who had helped him develop this new territory. Under the agreement the merchants promised not to rebel against the state, and Duke Huan pledged himself not to compel the merchants to sell, nor to seize their merchandise by force, nor to inquire into their capital or profits. This account indicates that, as early as the end of the ninth and beginning of the eighth century B. C., the importance of merchants in Chüng had won the recognition of the state and had secured for them an official protection not previously enjoyed.

The growing importance of commerce was even more manifest in the state of Ch'i, where the government itself engaged in trade. When Duke Huan (not to be confused with the ruler of Chüng with the same title) assumed its rule in 685 B. C., Ch'i was a very small state on the lower stream of the Chi River, which formed its western boundary. Its eastern boundary was less than ten miles from its capital, Lin-tzǔ (also a modern city).24 However, the state was situated on the coast, where fish abounded and where salt could be easily produced from the sea. Kuan Chung (d. 645 B. C.), the Duke's chief minister, realized the potentialities of these natural economic resources. He formulated and put into practice his policy of "creating profits through [the production and sale of] fish and salt."25 In co-ordination with this policy he devised a means of market control through regulation of supply and demand.26 By putting these measures into effect Duke Huan in a short time raised the "tiny Ch'i" to a position of hegemony within the Chou empire. As a reward to Kuan Chung, Duke Huan granted to him the revenue from taxes on trade.27 His benefit from this revenue made Kuan Chung, the minister of a feudal lord, "wealthier than the ruler of a state."28

The salt and fishing industries from which Ch'i of the seventh century B. C. derived so much power had, of course, developed long before this date. Likewise, there must have been an earlier export trade in fish and salt already developed which Duke Huan and Kuan Chung promoted and expanded with such great success. Evidence for this is found in the early history of Ch'i as related by Ssǔ-ma Ch'ien in his Shih-chi 29 The historian states that when T'ai-kung Wang, the first ancestor of the house of Ch'i, was enfeoffed and granted Ying-ch'iu, a town in Ch'i, he found that its soil was alkaline and its population small. Therefore, he "promoted its silk textile industry, perfected its skillful handicrafts, an dopened up [production and trade in] fish and salt."30 As a result, "both people and goods came to it [the town of Ch'i]. They arrived carrying babies on their backs and converged on it like the spokes of a wheel. Consequently, Ch'i provided the world with hats, sashes, clothes and slippers."31 Although the enfeudation of T'ai-kung Wang with Ch'i has been proven untrue, the remainder of the account may contain elements of truth.32 If so, Ch'i must have been an industrial center in ancient China for a long period with exports not only of sea products but also of handicrafts, particularly silks, which are mentioned by Ssǔ-ma Ch'ien elsewhere.33 Its immediate markets were the territories of Liang (Wei), Chao,34 Sung and Wei, according to Kuan-tzǔ.35 Corroboration of the great demand for the sea produce of Ch'i and of its eastern neighbor, Lai, is found in the gratitude of consumer states to Duke Huan when he abolished custom tolls and promoted direct purchases from Lai.36

Another fact which attests to early development of trade in Ch'i is the mention of "big merchants and hoarders"37 in conjunction with Kuan Chung's program for maintaining equilibrium of the market. Kuan Chung proposed to regulate trade and prevent market manipulation by storing ample stocks of grain in times of plenty for release in times of scarcity.38 His plan was called "well conceived and well adapted to problems of scarcity and oversupply."39 The regulation of the market reveals two significant points: trade played an essential part in Ch'i's economy and "big merchants and hoarders" had appeared who manipulated the market and "forcefully exploited the people."40 Even if we grant a measure of discount to the account of Kuan-tzǔ 41, the fact remains that trade had developed to an advanced stage in the Ch'i economy.42

We have evidence that by the middle of the seventh century B. C., and possibly earlier, economic needs had transcended state boundaries and that political divisions proved a hindrance to normal exchange. In 651 B. C. a conference of feudal lords convened at K'uei-ch'iu where Duke Huan of Ch'i demanded that the participants henceforth "shall not hoard grain" and "shall not curtail (the export of) raw materials."43 Because of its alkaline soil Ch'i was unable to support its entire population and the large army it required to maintain its hegemony. It is also probable that its handicraft industries needed raw materials from other areas. These were the reasons, no doubt, for the demands of the Duke. Another treaty drawn up in 562 B. C. practically repeats the provisions of the earlier one. In it the agreeing parties promise "not to hoard grain in bad years" and "not to block (the flow) of products."44

An early development of industry and commerce can be traced also in the state of Wei. In 658 B. C., the year after the state was invaded by the Ti people, Duke Wün moved his capital eastwards to Ch'u-ch'iu on the northern border of the present Honan province. There he pursued a program of reconstruction by "promoting commerce and favoring industry." As a result, the population of Wei increased threefold in a period of twenty-three years.45 More than a century later, when Confucius visited the state he was greatly impressed with its flourishing condition.46

Geographically speaking, Wei was situated at that time in the center of ancient China, in the plain at the middle of the old Yellow River valley, and on the Wu-tao (cross-roads).47 During the Ch'un-ch'iu period, Wei, capital of the state, was one of the three cities renowned for their riches, the other two being Lin-tzǔ, capital of Ch'i, and T'ao. It has on several occasions been mentioned together with T'ao as a place abounding in wealth.48

Across the northern and western borders of Wei we enter the territory of the state of Chin. Although it lagged behind Ch'i and Wei, there are signs of an early development of commerce there. Duke Wün (636—628 B. C.) of Chin, the first of its rulers to bring the state to a position of power, realized the benefits of trade and promoted it for the benefit of both the state and its people. On assumption of rule in 636 B. C., he "reduced duties at the passes, flattened the roads, opened up commerce, and lessened the burdens of the peasants ... in order to better the life of the people."49 As a result, in the middle of the sixth century B. C. we find that "the rich merchants of Chiang (capital of Chin) ... could decorate their carriages with gold and jade and have their clothing embroidered with flowery patterns."50 "They could," furthermore, "(befriend and) distribute gifts to the feudal lords."51 These words of Shu-hsiang, the grand tutor, to Han Hsüantzǔ, chief minister of state, give a good picture of the amount of wealth that merchants had accumulated in this state. From the degree of their prosperity we can easily infer the state of development of trade. As an ancient folk saying put it, "The longer the sleeves, the better the dancer dances; the wealthier the merchant, the more successfuly he trades."52

As a result of increased development in the seventh century B. C., trade was recognized to be as essential as agriculture and industry. A simultaneous and balanced development of the three became a criterion by which the strength of a state was judged. In 597 B. C., when Chin was preparing an attack on Ch'u, Sui-wu-tzǔ dissuaded Duke Li of Chin from acting, for as he observed, in Ch'u "neither the merchants, the farmers, nor the artisans have shown any relaxation in production."53 In 564, when Ch'u consented to join forces with Ch'in against Chin, Tzǔ-nang, a Ch'u minister, opposed the move, giving practically the identical reason.54 When, in 516 B. C., Duke Ching of Ch'i was concerned about the strength of his state, his minister Yen-tzǔ (named Ying), suggested that li be put into practice. Along with a few other administrative measures his li (proper principles for government) provided that "farmers do not shift their occupations, artisans and merchants do not change their professions. "55

Why did the ancient Chinese rulers consider the balance between trade, agriculture, and industry important? The Chou shu (Book of Chou) says, "If the farmers do not produce, there will be a shortage of food. If foresters do not produce, some works will not be accomplished. If the merchants do not produce, the sources of wealth will be cut."56 Ssǔ-ma Ch'ien remarks, "From the farmers food is obtained, by the foresters the materials are produced by which the artisans' goods are manufactured, by the merchants they are circulated."57 "These four," he continues, "are the sources for clothes and food for the people. If the sources are great, abundance will result; if the sources are small, scarcity will result. They enrich the state above, and enrich the people below."58

One way in which the state reaped direct benefit from commerce was through collection of taxes on market transactions and of customs duties at passes (kuan). To supervise markets and probably to collect taxes therein special officials were appointed. They were called ku-Chüng (director of trade) in Lu,59 ch'u-shih (market supervisor) in Chüng,60 shih-ling (market prefect) and shih-yüan (assistant to market prefect) in Ch'i,61 Ssǔ-shih (director of the market) and ku-shih (trade supervisor) in the Chou li.62

Frontier passes (kuan) as strategic points through which invaders might enter were originally the sites of military outposts. It was only after the volume of interstate trade had become substantial that they became also collection points of customs duties. The Kuan-tzǔ defines a pass as "a feudal lord's strategic road and the door for outside wealth."63 As soon as feudal rulers realized the large amount of revenue they were reaping from duty collections at frontier passes, greed inevitably led them to set up kuan for collection of duties within their interiors. Some states like Ch'i extended duty collections to points close to the walls of their capitals. In 522 B. C. Yen-tzǔ stated that "the passes [customs stations] close to the capital [of Ch'i] tyrannically collect duties for personal profit,"64 that is, of the prince of Ch'i. One hundred and three years before (in 625 B. C.) the state of Lu had set up six additional such customs offices. Confucius referred to their establishment as one of the three inhuman acts performed by the Lu minister, Tsang Wün-chung,65 The phraseology of Confucius gives the impression that these customs stations were in the interior. The whole practice of customs collections was bitterly denounced by Mencius: "In ancient times the erection of customs stations was directed against tyranny [meaning agression]; at the present [fourth century B. C.] the erection of customs stations is for exercising tyranny."66

The above data concerning the establishment of stations for collection of duties throw considerable light on the development of commerce. The lack of concrete figures or even general statements regarding the amount of revenue collected at a customs station makes it impossible to estimate the volume of trade. A portion of the Tso chuan, however, does enable us to gauge its value. The author of the Tso chuan relates that the Ti people invaded Sung during the reign of Duke Wu (766—749 B. C.). In the battle with the invaders all the Sung generals perished except Erh-pan, whose chariot led the defense's charge. To reward him, 'The Duke granted a customs station to Erh-pan as his fief and let him live on its collections."67 That the collection of duties was granted as a reward or fief to a victorious warrior indicates that this source of revenue had become sizeable and fairly regular. Even more significant is the fact that Erh-pan's grant took place in the middle of the eighth century, almost a hundred years before the commercial policies of Duke Huan of Ch'i and Duke Wün of Wei were adopted.

The development of trade and the large profits accruing to those engaged in it would naturally bring merchants out of relative obscurity into activity in public affairs. The names of a few merchants before the fifth century B. C. are mentioned in extant historical literature in connection with important events. Primary among these are the names of Pao Shu-ya and Kuan Chung, whom we have already mentioned as the advisor of Duke Huan of Ch'i.

Before his rise to prominence, or in his own words "when I was in a difficult situation," Kuan Chung had been a merchant, originally from Ying-shang68 (in what is now Central Honan), trading in Nan-yang,69 a large city in the southwestern part of the province. For sometime he had been a business associate of Pao Shu-ya,70 who likewise rose to high position in the ruling circle of Ch'i. It was, in fact, Pao Shu-ya who recommended Kuan Chung to Duke Huan. Both Pao Shu-ya and Kuan Chung became high officials in the state. The latter was honored by Duke Huan with the title chung-fu 71 and Confucius paid tribute to him for having saved the Chinese from conquest by barbarians.72

Next we find the name of Hsüan Kao, a merchant of Chüng.73 In 627 B. C., when driving his herds of cattle to the city of Chou to market them, Hsüan Kao met the armies of Ch'in marching eastward to make a surprise attack on his home state. Sensing the danger, he pretended he was an official emissary sent by Chüng to welcome and feast the invading troops. While entertaining them he secretly dispatched warning and thereby saved his state.74

From Chüng there was another merchant, whose identity is unknown. In 568 B. C., an important general of Chin, Hsün Ying, was captured by the enemy in the course of a battle with the army of Ch'u. Partisans of the general asked a merchant, who had come to Ch'u on business, for help in an escape plot. The merchant agreed and worked out a plan for smuggling the general out among his merchandise. Although Hsün Ying was released before the plot was carried out, he was nonetheless so grateful that when the merchant came to Chin to trade he offered him special favors. Declining the general's generosity, the merchant went on to Ch'i in pursuance of his business.75 Obviously this merchant must have been prominent and one with social connections among important personages in both Ch'u and Chin. To warrant such extensive travelling in Chüng, Ch'u, Chin and Ch'i, practically all over the then known Chinese world, his business must have been on a large scale.

Of all the big merchants of this period the most famous was Tuan-mu Ssǔ, a disciple of Confucius, better known as Tzǔ-kung. Tzǔ-kung, a native of Wei which had long prospered through trade, is said to have been a master of market manipulation. "He hoarded merchandise or released it according to the prospects of making profit,"76 and thereby acquired a great fortune. Although Confucius reproached him for his interest in trade and for his lack of it in studies, he praised his ability in commercial speculation.77 Tzǔ-kung's wealth enabled him to travel from one princely court to another accompanied by a long retinue of horses and carriages laden with fine silks of which he made gifts to the feudal princes. Wherever he went, rulers accorded him the courteous treatment of an equal.78 In the opinion of Ssǔ-ma Ch'ien, the great historian, it was Tzǔ-kung who "made the name of Confucius popular over the world."79 Even though he may not have done so consciously, Tzǔ-kung, as both the favorite disciple of Confucius and a prominent merchant, must have publicized his teacher widely. Moreover, he was not only a business man renowned for his wealth, but was an influential politician as well. While a trader he served Lu and Wei in various official capacities.80 His last years were spent in Ch'i, the state most famous for its industry and commerce.

Another personage reported to have come to Ch'i was the famous statesman Fan Li, who had aided the king of Yüeh (modern Chekiang) to conquer Wu (southern Kiangsu), to extend the state's territory to the borders of Lu and Ch'i, and thus to attain a political position on a par with the central states. This was in the first half of the fifth century B. C. According to the account, after Yüeh had attained its greatest power, Fan Li resigned, changed his name, went to Ch'i, and later established himself as a business man in T'ao, which city was regarded as the geographical center of the empire. Henceforth, Fan Li became known as T'ao Chu-kung (Old Gentleman Chu of T'ao). The fortune he made from trade profits and interest on money-lending became so great that he became a symbol of wealth81 and served as a model to Chinese businessmen from that day to the present. Though the identification of Chu-Kung with Fan Li seems open to doubt82 the historical character of the Old Gentleman Chu remains.

Like the Old Gentleman Chu of T'ao, Po Kuei also achieved great success in trade and gained even greater fame. He was a native of Chou, an area which was highly commercialized. Ssǔ-ma Ch'ien says that he lived in the time of Marquis Wün (446—397 B. C.) of Wei, but some modern scholars believe he lived a century later.83 According to Ssǔ-ma Ch'ien he "was very successful in predicting the trend of the time." Hoarding merchandise which would bring him large returns, he purchased when others dumped, and vice versa. "In grasping the advantages of the moment, he acted as ferocious animals and vultures do in falling on their prey."84 He boasted that he conducted his business in the manner I Yin and Lü Shang (statesmen of an earlier time) laid out their administrative policies, Sun Pin and Wu Ch'i (famous military strategists) commanded their armies, and Shang Yang (a reformist statesman) executed his orders.85

The above accounts of early Chinese merchants are not to be read as biographical notes only, for in them we find data upon which a clearer picture of early commerce in China can be reconstructed. Fragmentary as the information is, it all points to a considerable development of trade in this period. Let us summarize our findings. As far back as the early part of the eighth century B. C. the contribution of commerce to general economic life had won the attention of the ruling authorities. Around the middle of that century custom duties collections, at least in Sung, had reached considerable proportions. In the following century Ch'i and Wei successively pursued programs of commercial expansion. The lucrative profits of trade attracted many to take it up as an occupation even though it had been considered an ignoble one. Merchants travelled throughout the the world then known to the Chinese and amassed such fortunes that nobles accepted them as equals and appointed them to high administrative positions in their governments.

End Notes

22 Tso chuan, Ssǔ-pu pei-yao ed., XLVII, 8b —10.
23 Originally the territory of the state of Chüng was in the area below the Wei River in eastern Shensi.
24 Kuo-yü (Stories of the States), Ssǔ-pu pei-yao ed., VII, 9a.
25 Shih-chi, XXXII, 8b, and XXX, 20b.
26 Ibid, and op. cit., CXXIX, 2b.
27 The term for the market tax is san-kuei image When Confucius was asked whether Kuan Chung was frugal, he said he was not because Kuan Chung "had the san-kuei." (Lun-yü, III, 5b). In the Han-fei-tzǔ it is said after Kuan Chung became the chief minister of Duke Huan of Ch'i, the latter bestowed upon him the san-kuei in order to enrich him. (1875, XII, 11a). As the term was obscure to later scholars, it has been interpreted as meaning "the name of a terrace" or "wives from three different families." Kuo Sung-t'ao image (1818 —1891) rejects these explanations and suggests that it was a general term applied to market taxation, meaning thirty per cent of the profit. See his Yang-chih shu-wu wün-chi (A collection of writings of the Yang-chih Study), I, "Shih san-kuei" (Interpretation of San-kuei). Kuo Sung-t'ao's interpretation appears most satisfactory. The san-kuei grant to Kuan Chung as recorded by Liu Hsiang image (77 —6 B. C.) is "one year's tax from the market of the Ch'i state." (Shuo-yüan, Ssǔ-pu pei-yao ed., VIII, 12a). The market from which Kuan Chung was to receive his revenue was probably that in the Ch'i capital.
28 Shih-chi, CXXIX, 2b. In LXII, 3a, Ssǔ-ma Ch'ien gives a slightly different statement of Kuan Chung's wealth. There he says, "Kuan Chung was so wealthy that he equalled the house of the ruler of state (of Ch'i)."
29 The story quoted below in the text concerns the enfeudation of Lü Wang image or T'ai-kung Wang (the first ancestor Wang) with Ying-ch'iu image a town in the ancient Ch'i state, as his fief. It tells how Lü Wang established his state in Ch'i and developed its economic resources. The story does not correspond with the historical facts. As has been ably disproved by Prof. Fu Ssǔ-nien, at the time when Lü Wang was supposed to have been made the feudal lord of Ch'i, the territory which later came to be known as Ch'i was still in the hands of the Shang people or their vassals. The very name of the beneficiary, Lü Wang or Wang of Lü, indicates strongly that the fief of Wang was Lü, not Ch'i. Even a generation later, his son Chi image was still called Lü Chi or Chi of Lü. Prof. Fu's arguments are contained in his article "On Ta-Tung and Hsiao-Tung," Bulletin of the National Research Institute of History and Philology, Academia Sinica, II (1930), 101 —109. What immediately concerns us here is not the authenticity of the enfeudation of Lü Wang in Ch'i but the possibility of the early development of industry and commerce in the region of Ch'i, which is the main point of the story.
30 Shih-chi, CXXIX, 2b.
31 Ibid. This statement and the one immediately preceding are significantly absent in Shih-chi, XXX, 20b, where the historical development of industry and commerce is related by Ssǔ-ma Ch'ien and in LXII where the history of the state of Ch'i is treated by the same historian. In neither place is the first ancestor of the house of Ch'i mentioned to be the first developer of the economy of Ch'i. This is another point which can be used to support Prof. Fu in his argument disputing Lü Wang as the first ruler of the feudatory Ch'i. The story about Tai-kung Wang seems to have some historical elements with its features borrowed from the story about Duke Huan and Kuan Chung.
32 See note 29.
33 Such as silk fabrics recorded in Shih-chi, CXXIX, 10a. Li Ssǔ speaks of the fine kao of O image as one of the luxurious industrial goods imported into Ch'in in his memorial to the king of Ch'in quoted above. O was a town in Ch'i, located northwest of the modern city of Tung-o in western Shantung.
34 Neither Liang (Wei) nor Chao, the state mentioned immediately after Liang, existed during the Ch'un-ch'iu period which we are discussing. These states came into being only after 431 B. C. through a split of the territory of Chin. The creation of these states received official sanction in 403 B. C. when the king of Chou granted the rulers of the two de facto states the status of the hou or marquis. We use the names of the two states instead of Chin to preserve the original wording of the passage in Kuan-tzǔ, which is here referred to.In the Chou period, there were two states whose name was pronounced Wei. One, written image existed throughout the whole Chou period, and the other, written image was officially created in 403 B. C. Although the names of these two states cannot be confused in Chinese script, they can be easily in English. In order to avoid the confusion, we will refer to the state created in 403 B. C. as Liang, which was the name of its capital. In the literature of the Chan-kuo period this state is frequently so designated.
35 Kuan-tzǔ, XXIII, 15b.
36 Kuo-yü, VI, 10b.
37 Op. cit., XXII, 6b. The statement is also quoted in Han shu, i. e. Ch'ien Han shu 1641 ed., XXIV, Part 2, 1b.
38 Ibid.
39 Shih-chi, XXX, 20b.
40 Kuan-tzǔ, XXII, 6b.
41 The Kuan-tzǔ, attributed to Kuan Chung, is a work of the Ch'an-kuo period (403 — 221 B. C.) which contains many later interpolations. However badly interpolated, it does contain valid Ch'i traditions. The chapter here quoted is mentioned by Ssǔ-ma Ch'ien as also a part of the work as circulated at his time (145 —86 B. C.?). Moreover, the wording of Pan Ku's (32 —92) quotation (XXIV, Part 2, 1a —1b) of the portion relating to market regulation is identical with the present text.
42 Huan K'uan image of the later Han dynasty (25—220) speaks of the Ch'i commercial caravan consisting of three thousand carriages. Yen t'ieh lun, Ssǔ-pu pei-yao ed., II, 6a.
43 Müng-tzǔ, Ssǔ-pu pei-yao ed., XII, 9b; Tso chuan, XIII, 4b —5a.
44 Tso chuan XXXI, 10.
45 Op. cit., XI, 8b.
46 Lun-yü, XII, 3a —3b.
47 The interpretation of wu-tao image as meaning "cross-road" is advanced by Müng Wün-t'ung image in his article "Lun ku shui-tao yü chiao-t'ung," (A Discussion of Ancient Rivers and Communication), Yü-kung (Chinese Historical Geography), II (1935), No. 3, p. 4.
48 For more information on the commercial centers of T'ao and Wei see below pp. 46 —47.
49 Kuo-yü, X, 13b.
50 Kuo-yü, XIV, 11a.
51 Ibid.
52 Han-fei-tzǔ, 1875, XIX, 10a.
53 Tso chuan, XXIII, 3a.
54 Tso chuan, XXX, 15a.
55 Tso chuan, LII, 7a.
56 Shih-chi, CXXIX, 2a.
57 Ibid.
58 Ibid.
59 Tso chuan, LI, 12b.
60 Tso chuan, XXXII, 3a.
61 T'ien Tan image the famous general of Ch'i is said to have previously held the position of the shih-yüan image of Lin-tzǔ. Shih-chi, LXXXII, 1a. In the ancient Chinese official hierarchy yüan was assistant to the ling (prefect) or chang (chief) of an office. Since there was the position of yüan there must also have been the position of ling or chang.
62 Chou li, Ssǔ-pu pei-yao ed., XIV, 7b and XV, 2b. (Hsün-tzǔ, 1876, XV, 8b, has also ku-shih but it is not certain whether this is a title of an official or a general term to mean "teacher of merchants.").
63 Kuan-tzǔ, IX, 15b.
64 Tso chuan, XLIX, 7b.
65 Tso chuan, XVIII, 8a.
66 Müng-tzǔ, XIV, 3a —3b.
67 Tso chuan, XIX, Part 2, 2a.
68 Shih-chi, LXII,1b.
69 Lü-shih ch'un-ch'iu, quoted by Ssǔ-ma Chüng image a T'ang commentator of the Shih-chi (LXII, 1a).
70 Shih-chi, LXII, 1b.
71 Meaning next to one's father in honor.
72 Lun-yü, XIV, 5b.
73 Lü-shih ch'un-ch'iu records a partner of Hsüan Kao image by the name of His shih image 1875, XVI, 12a.
74 Tso chuan XVII, 7b —8a. The story is slightly differently worded in Shih-chi, V, 14b —15a.
75 Tso chuan, XXVI, 3b.
76 Shih-chi, LXVII, 12a.
77 Lun-yü, XI, 4b.
78 Shih-chi, CXXIX, 5a.
79 Ibid.
80 Tso chuan, LVIII, 4a; LIX, 2a; 12a. Shih-chi, CXXIX,5a.
81 Shih-chi, CXXIX, 4b —5a.
82 See Ch'ien Mu image Hsien-Ch'in chu-tzǔ hsi-nien k'ao-pien (A Study of the Chronology of the Pre-Ch'in Philosophers), Shanghai, 1935, 101.
83 Shih-chi, CXXIX, 5a. For the critical discussion on the life-time of Po Kuei see Ch'ien Mu, op. cit. 234 —236.
84 Shih-chi, CXXIX, 5a —5b.
85 Shih-chi, CXXIX, 5b.


The Chan-kuo Period (403—221 B. C.) witnessed a further development of commerce. The civil wars regarded as characteristic of the period (Chan-kuo means "warring states") in actuality surpassed very little those of earlier times in either frequency or destructiveness. They certainly do not appear to have hindered the growth of trade.

It is significant that the agrarianist Hsü Hsing, a contemporary of Mencius (390—305 B. C.), advanced a political philosophy which aimed to eradicate the evils of the day by making a farmer of everyone. It is said that he and his disciples cultivated fields themselves and lived extremely simple lives so as to set an example for their teachings. Mencius disapproved of both their theory and their practice of it. Encountering a follower of Hsü Hsing he questioned him and learned that the philosopher had been unable to maintain himself without acquiring clothes, hats, utensils and iron implements from other people. As Mencius puts it Hsü Hsing was engaged in a "busy and confused" [i. e., 'complicated'] exchange with the hundred [i. e., 'many'] artisans for the goods [he needs]."86

Some of the things which Hsü Hsing needed could be purchased in the locality (both he and Mencius lived in the state of T'üing at the time), such as simple pottery utensils. Some others, such as iron for making tools, could not be so obtained. Artisans who manufactured iron tools had to secure their metal from other areas.87 Many other things were as necessary to life on an economically higher level as iron was to the ascetic. Timber, bamboo, ku barks for writing material, lu mountain hemp for making cloth, yak tails, jade and other precious stones from west of the mountains;88 fish, salt, lacquer, silk, musical instruments and embroideries89 from east of the mountains; wood of the nan (Machilus nanmu) and the tzǔ (Lindera tzumu), ginger, cinnamon trees, gold, tin, lead, cinnabar, rhinoceros (hide or horn), tortoise shells, pearls, ivory, and other hides from south of the Chiang (Yangtze River); horses, oxen, sheep, furs, sinews and horns from the north; and copper and iron from many other places90 — "all these," as Ssǔ-ma Ch'ien observes, "were what the people of the central kingdom91 enjoyed and the materials from which, according to the custom of the day, were made clothes, food and articles for nurturing the living and burying the dead."92

These commodities listed above are taken from the introduction of the section on merchants and manufactures in pre-Ch'in China in Shih-chi by Ssǔ-ma Ch'ien. What he describes is but a general picture,93 and he does not pretend to give a complete enumeration. Besides these, jujubes from Yen (roughly modern Hopeh) and An-i (southern Shansi), chestnuts from Yen and Ch'in (roughly Shensi), fish and salt from Yen and Wu (southern Kiangsu), copper from Wu, copper and iron from Pa and Shu (Szechuan), fruits and cloth from Pan-yü (Canton), oranges from Shu, Han and Chiang-ling (Szechuan and southern Hupeh), etc., formed part of the merchandise which crowded many of the markets.94

Flourishing trade brought about commercialization of a part of the agricultural produce. We find it stated that a cattle breeder who possessed 50 horses, 166 oxen, 250 sheep, and 250 pigs; a fish grower who produced 1000 piculs (shih) in his ponds; an orchardist of 1000 jujube trees in An-i, or of 1000 chestnut trees in Yen or Ch'in, or of 1000 orange trees in Shu, Han or Chiang-ling; a grower of 100 ch'iu trees (Mallotus japonicus) in the Yellow River valley, or of 1000 mou (land measure) of lacquer trees in Ch'ün or Hsia (central and eastern Honan), or of 1000 mou of mulberry trees or hemp in Ch'i or Lu (Shantung); or of 1000 fertile mou of grain, or of 1000 mou of the chih and the ch'ien plants (from the flowers of which red and yellowish-red pigments were made), or of 1000 plots of ginger or leeks — that any one of them received a revenue equal to that of a marquis with a fief of one thousand households.95 Since a household paid an annual tribute of two hundred cash to its noble lord, this revenue in terms of cash would be 200,000. Calculated at the twenty per cent rate of profit accruing to the farmers, artisans and merchants of the time as recorded by Ssǔ-ma Ch'ien, a profit of 200,000 represented, so to speak, a capital of one million.96 Ssǔ-ma Ch'ien therefore calls the man of such wealth a "noble without a fief."97

According to the same historian the following merchandise was sold yearly in a large city: 1000 jars of wine, 1000 long-necked jars of vinegar, 1000 big jars of sauce, 1000 heads of butchered oxen and cows, sheep or pigs, 1000 chung (1 chung = 64 tou or Chinese pecks) of grain, 1000 wagonloads of fuel, boats of a total length of 1000 chang (1 chang = 10 ch'ih or Chinese feet), 1000 pieces of lumber, 10,000 bamboo poles, 100 carts, 1000 ox wagons, 1000 pieces of lacquered furniture, 1000 chün (1 chün = 30 chin or Chinese catties) of bronze vessels, 1000 shih (1 shih = 120 catties) of plain furniture, 1000 shih of iron utensils, 1000 shih of chih and ch'ien yellowish red pigment, 200 horses, 250 oxen and cows, 2000 sheep, 2000 pigs, 100 slaves, 1000 chin (Chinese catty) of animal sinew, horn and cinnabar, 1000 chün of silk and fine cloth, 1000 bolts of embroidered silk, 1000 shih of t'a-pu (cotton cloth) and hide, 1000 tou (pecks) of lacquer, 1000 chin of t'ai fish (a sea fish) and knife fish, 1000 shih of Chou (miscellaneous small) fish, 1000 chün of pao fish, 3000 shih of jujube and chestnuts, 1000 fox and seal furs, 1000 shih of lamb skins, 1000 felt rugs, and 1000 chung of fruits.98

Ssǔ-ma Ch'ien's record, being given in round numbers, cannot be taken as accurate in detail, nor can it be assumed that it is free from the exaggeration of over-enthusiasm. Nevertheless, the bulk of the statistics he gives, inaccurate though they may be, does reflect the large volume of business transactions in a sizeable city of that day. Again according to Ssǔ-ma Ch'ien, a merchant could reap a profit of fifty per cent, though some might make less.99 In pursuit of trade, the merchants came and departed in large noisy crowds.100 Profit from their trade was so great that Ssǔ-ma Ch'ien observed, "For a poor man seeking wealth, farming was not as good as handicrafts, and handicrafts not as good as trade; to embroider silk was not as good as to lean against a shop door."101 "Once a man became rich, all virtues would attach to him."102 "Precedence was accorded to wealth, and propriety and modesty were not given highest attention."103

During this period, many opulent merchants seem to have arisen to considerable influence. In the middle of the third century B. C. Lü Pu-wei, an influential merchant from Yang-ti in the state of Han, interfered with the line of succession in Ch'in. Mainly through bribery he helped onto the throne both King Chuang-hsiang (249—247 B. C.) and later, King Chüng (246 B. C.), who unified China and became known as its First Emperor (221—210 B. C.). Lü Pu-wei was appointed Chancellor (ch'üng-hsiang) by King Chuang-hsiang, and received the title of Marquis of Wün-hsin carrying with it a fief of 10,000 households in Lo-yang. When King Chüng ascended the throne, he further honored him by making him the Chancellor of the State (hsiang-kuo) and by calling him chung-fu, meaning "next to his father." With his wealth he kept ten thousand slaves and three thousand "guests" ( clients in the sense given to the word in Roman history). Among his clients were a number of scholars who at his order composed the historically famous work, the Lü-shih ch'un-ch'iu or the Lü lan (Book of the Lü).104

Another result of the development of trade during the Chan-kuo period was the rise of the "metropolitan centers" (tu-hui). Upon them "the people from the four directions" converged, and from them merchandise flowed to distant corners of the continent of Asia.

In the royal domain of Chou, Lo-yang was the commercial metropolis. The people of Chou were well-known artisans and tradesmen, who received a return on their output, as Ssǔ-ma Ch'ien gives it, of twenty per cent profit.105 The rich boasted of their ability to go on business trips of long periods, and the poorer imitated the rich.106 Through them their metropolis traded with Ch'i and Lu in the east, with Liang (Wei) and Ch'u in the south.107 "Situated in the middle of Ch'i, Ch'in, Ch'u and Chao, it was (as if) on a street."108

In the territory of Ch'i, specifically "between the sea and Mount T'ai (in central Shantung)," Lin-tzǔ was the trading center.109 Besides being the capital of the state of Ch'i, Lin-tzǔ was a commercial city with a long history and unmatched prosperity (see below p. 177). Through the four centuries until King Hsüan (319—301 B. C.) of Ch'i, Lin-tzǔ was perhaps the richest and most prosperous city in the world then known to the Chinese. As Su Ch'in described it to the King, "On the roads to Lin-tzǔ the wheels of carriages bump each other, and the shoulders of the people rub one another. Connected, the breasts of their coats form a curtain; lifted, their sleeves form a tent; swept, their sweat makes rain."110 It was an emporium of salt, fish, and various silk fabrics.111

The metropolitan center in the northeastern part of ancient China was Chi, capital of the state of Yen. It was economically connected with Ch'i and Chao in the south, and traded with the barbarian peoples to its north and east beyond the Chinese borders.112 It provided the interior of ancient China with animal furs, sinews, horns, and horses.

In Chao, the commercial center was Han-tan. It was the state's capital, a center of iron production and many other skillful handicrafts, and an economic pivot for the regions between the Chang and the Yellow River valleys. Its merchants frequented Chi and Cho in Yen in the north and the areas of Chüng and Wei in the south.113 Though Ssǔ-ma Ch'ien does not so state, there certainly must have been constant visitors on business from Ch'i in the east and from the regions beyond the T'ai-hang mountain range in the northwest. Yang and P'ing-yang, cities in the Fün River valley in central Shansi, were also points of trade. They were the hubs for trade with Ch'in and the Ti barbarians in the west and with Chung and Tai in the north.114

Similarly, Wün and Chih constituted the commercial centers south of the T'ai-hang Mountains and north of the Yellow River in present day northwestern Honan. These two cities traded with Shang-tang in the west and with Chung-shan in the state of Chao, in which localities the land was not fertile and the population so large that large numbers were forced to take up trade or skilled handicrafts to earn their living.115

Kuan-chung in modern central Shensi was another important economic area, in which Li-i, replete with "big merchants," served as a commercial center in northwestern China. It supplied the east with the goods from the non-Chinese peoples in the north and the west. To the south it controlled the merchandising of the special products of Pa and Shu (modern Szechuan): the chih deep red pigment, ginger, cinnabar, precious stones, copper, iron, and utensils made of bamboo and wood, and monopolized the importation of the slaves from P'o, an area neighboring Pa and Shu in their south. It also had access to the horses and yaks of the barbarians beyond the southwestern Chinese border.116

After the Yangtze River emerged from the gorges on the eastern border of Pa, the first big city on its banks was Ying (modern Chiangling), the capital of Ch'u. Ch'u was famous for its economic resources from the Yün-müng,117 an extensive area which, according to Wang-sun Wei, a Ch'u minister, produced metal (gold), lumber, bamboos, tortoises, pearls, furs (of wild animals), hides (of rhinoceros), feathers, and (yak) tails.118 In the lower Yangtze (called Chiang at that time) Valley, we find the city of Wu, still known today by its old name, which Ssǔ-ma Ch'ien calls "the metropolitan center east of the Chiang." The salt from the sea, the copper from the Chang-shan mountains, and the products from the many rivers and lakes of this area constituted the main merchandise of this city for export.119

The center north of the Huai River was Shou-ch'un, the last capital of the Ch'u. It lay at the juncture of communication routes between the Yangtze Valley and the Yellow River Valley. It was an emporium for animal hides and timber120 as well as many other southern products destined for the northern regions.

In ancient times the region between the Huai and the Yellow Rivers corresponding to what is now northern Kiangsu, eastern Honan, northern Anhui and southern Shantung, was traversed by a number of small streams and artificial canals. The most important of the canals was the Hung Kou or the Great Canal. The date of its construction is not exactly known. The description of the Great Canal in the treatise on rivers by Ssǔ-ma Ch'ien immediately follows the legend of Yü, who was said to have ended the Great Flood by opening nine water-ways in the lower valley of the Yellow River, and precedes the account of the construction of the canals of the Ch'un-ch'iu period (770—481 B. C.). At the latest, the canal appears to have been in existence before the fifth century B. C. Its starting point was in Ying-Yang (same as the modern district of the same name in central Honan) where it connected with the Yellow River. It ran eastwards parallel with the present Lunghai Railway, passed Liang (modern Kaifeng), capital of Liang (Wei), then turned its course southeastwards, and joined the tributaries of the Huai River in modern northwest Anhui.121 As Ssǔ-ma Ch'ien sees it, the construction of the canal "was to open the way to and connect the states of Sung, Chüng, Ch'ün, Ts'ai, Ts'ao, and Wei, and to join the Chi, the Ju, the Huai, and the Ssǔ Rivers."122 These states covered a large area of ancient east central China, corresponding to what is today northern Honan, southern Hopeh, southwestern Shantung, central and eastern Honan, northwestern Anhui, and northern Kiangsu. The four rivers mentioned were the main streams of eastern and southeastern China in ancient times.

Another canal whose importance in the economic life as well as in the political struggles of this period was the Han Kou or the Han Canal. This canal was constructed in 486 B. C. by the state of Wu.123 It drew its water from the Chiang (Yangtze River) below the city of Chiang-tu, an historical commercial center. It followed a northward course and joined the Shü-yang Lake in modern central Kiangsu. Emerging from the lake it continued its northward course and joined the Huai River in the area of modern Huai-an. In the Huai-an area it was also connected with the I River which flowed into southern Shantung and the Ssǔ River which ran to the northwest and met the Chi River.124 When the Sui dynasty (589—617 A. D.) constructed the Grand Canal, which was to prove so valuable to later dynasties for transportation of rice from the south, it made use of the original course of the Han Canal for the middle of the present Grand Canal course.

Thus we see that during the later part of the Chou dynasty, the great plain north of the Yangtze, south of Mount T'ai (in central Shantung), west of Lo-yang, capital of Chou, extending as far as the sea, traversed by many water ways, had become an economic whole, in spite of political boundaries. In the heart of the water communication system were situated the states of Sung with its capital Shang-ch'iu located on the site of its modern namesake in eastern Honan, and later P'üng-ch'üng which was what is now the city of T'ung-shan or SuChou in northern Kiangsu, and Liang with its capital Ta-Liang located on the Great Canal itself. Keeping in mind the strategical location which Sung occupied we will easily and fully understand the reason why during the Ch'un-ch'iu period this small state became a prey of Chin, Ch'u and other powers. We will also easily understand how the newly created state of Liang, relying on its strategic position, could maintain a hegemony for almost a century from 425 to 334 B. C.125

In this great plain knit by the well laid water-ways of communication and teeming with commercial activities there were a few cities which may be regarded as "metropolitan centers" in the sense of the word as Ssǔ-ma Ch'ien uses it. But the most famous of them all was T'ao, where the Old Gentleman Chu made his fortune. The city, located in modern Ting-t'ao County in southwestern Shantung, was on the bank of the ancient Southern Chi River. It was situated midway between Lin-tzǔ, Lo-yang, Han-tan and Shou-ch'un — four great metropolises which we have described above. By land route it was within easy access to Han-tan. By the Chi River it could reach both Lin-tzǔ126 and Lo-yang. The Great Canal and the Chi River connected it with Shou-ch'un. It was on the Wu-tao (cross-road) of ancient China 127 and regarded as the richest place which could be matched only by Wei.128 It was, as Ssǔ-ma Ch'ien puts it, "the center of the world, from where all the feudal states could be reached and where merchandise was bought and sold."129 Hence it was coveted by all the warring states. In 386 B. C. the Chao state moved its capital to Han-tan, and in 362 B. C. Liang moved its capital to Ta-Liang. The action of both states was aimed at the struggle for the "Cross-road" and the area around the metropolis of T'ao. When the state of Ch'in embarked on its conquest of the rest of the Chou empire it first cut a long corridor through the territories of Han and Liang and took T'ao before 291 B. C., seventy years before it accomplished the conquest. The occupation of T'ao, the economic center and strategic point of the day, must have contributed much to the unification of ancient China by Ch'in in the following decades.

The rise of T'ao and the importance of the "Cross-road" were direct results of the commercial development during the Chan-kuo period. The economic forces of trade also created other metropolises, which served as focal points for their respective regions and helped bring about political unification.

However, from all this it is not to be understood that the economic life of China before the end of the third century B. C. had become highly commercialized. China was, as it still is generally, fundamentally agrarian. Economic production was not as much for the market as it was for the immediate needs of the household. Yet, in the framework of this agrarian economy, commerce had made its appearance, grown in importance, served well the economic life of ancient China, and called forth the use of metallic money, which gained increasing significance as trade developed ever further.

Even if the use of metallic money had been prevailing, it may be assumed that not every business transaction was made through this medium of exchange. Compared with Chou China, the Mediterranean world of antiquity appears to have been more commercialized. Even in this case Prof. W. L. Westermann has cautioned us not to minimize exchange in kind against exchange in money; for, he argues, "Constantly throughout antiquity exchange in natura and exchange in money form appear side by side."130 The same may be also said of China of the Chou period.

End Notes

86 Müng-tzǔ, Ssǔ-pu pei-yao ed., V, 9a —11b.
87 T'üng is not known as an iron producing area in either ancient or modern times. According to Ssǔ-ma Ch'ien, places where iron industry produced great fortunes in the Chan-kuo period were Chao including Han-tan, Lin-Ch'iung in Shu, Liang (Wei), Wan in Ch'u and Lu. For the iron smelter Kuo Tsung of Han-tan see Shih-chi CXXIX, 66; for the iron smelter of the Cho family of Chao and later of Lin-ch'iung in Shu, see ibid., 17a; for the smelter of the Kung family of Liang and later of Wan in Ch'u, and also for the smelter family of Ping of Lu, see ibid., 18a. As Ssǔ-ma Ch'ien mentions only iron manufacturers of great wealth, this does not preclude the existence of relatively small producers in still other districts.
88 The Chinese term is Shan-tung, meaning "east of the mountains." The mountains referred to here must be the range of T'ai-hang which spreads over the central region of ancient Chinese civilization on the border of modern Shansi and Honan Provinces. However, the term which was used obviously in a general sense should not be understood literally. As Wang Ming-shüng image has pointed out, during the Chan-kuo period it denoted the regions east of the Han-ku Pass in western Honan, roughly east of the state of Ch'in. See his Shih-ch'i-shih shang-ch'iao (Discussions on the Seventeen Dynastic Histories), 1667, XXXV, 1a —2b.
89 The word used by Ssǔ-ma Ch'ien is image meaning "colors" or "colorful things." Embroideries being colorful fabrics, we may assume that they are what the historian meant. In Shih-chi CXXIX, 10a, where the noted local products of the state of Ch'i are mentioned, Ssǔ-ma Ch'ien does mention embroideries using the term wün-ts'ai image which signifies "patterned and variegated fabrics."
90 See note 87.
91 The term chung-kuo, generally rendered as "central kingdom," denoted the interior of the country as distinguished from the vassal states in the border regions, and China as distinguished from non-Chinese peoples.
92 Shih-chi, CXXIX, 1b.
93 Ibid.
94 Shih-chi, CXXIX, 7b, 9b, 11a, 12a, 15a.
95 Shih-chi, CXXIX, 14b—15a.
96 Shih-chi, CXXIX, 14b.
97 Shih-chi, CXXIX, 14a.
98 Shih-chi, CXXIX, 15b —17a. In enumerating quantities of merchandise we have altered the wording of the Shih-chi text from which we have quoted. For instance, in the original Chinese the phrase "100 slaves" reads t'ung shou ch'ih ch'ien (one thousand fingers of slaves). Since each slave has ten fingers, we have changed the wording to a simpler expression. Similar changes have been made at other points.
99 Shih-chi, CXXIX, 17a.
100 Shih-chi, CXXIX, 3a.
101 Shih-chi, CXXIX, 15b.
102 Shih-chi, CXXIX, 3a.
103 Shih-chi, XXX, 20b.
104 For the life story of Lü Pu-wei see his biography in Shih-chi, LXXXV, 1 ff. Cf. also Derk Bodde, Li Ssǔ.
105 Shih-chi, LXIX,1a.
106 Shih-chi, CXXIX, 18b.
107 Shih-chi, CXXIX, 9b.
108 Shih-chi, CXXIX, 18b.
109 Shih-chi, CXXIX, 10a.
110 Chan-kuo ts'ü, Ssǔ-pu pei-yao ed., VIII, 8b.
111 Shih-chi CXXIX, 1a and 2b.
112 Ibid.
113 Shih-chi, CXXIX, 8b.
114 Shih-chi, CXXIX, 9a. Ssǔ-ma Ch'ien's exact statement reads: "(The people) made a living by relying on the profits from skillful works." Here the "skillful works" may include both artifice handicrafts and trade. In the Yen t'ieh lun (Discourses on Salt and Iron) Chung-shan of Chao is described as a thoroughfare of the Chou empire, where "merchants infested the roads", and the people "took great interest in the secondary [meaning trade and handicrafts], enjoyed luxury, and did not devote themselves to the fundamental [meaning farming]. Their fields were not cultivated."(I,7a.)
115 Shih-chi, CXXIX, 7b.
116 Shih-chi, CXXIX, 10b —11a.
117 Kuo-yü, XVIII, 8a.
118 Shih-chi, CXXIX, 11a.
119 Shih-chi, CXXIX, 11b.
120 Ku Tsu-yü image (1624 —1690) holds that the main course of the Hung-kou system was dug by King Yen of Hsü image (Ching-shih ta-wün, in Chieh-i-t'ing chi, Ssǔ-pu ts'ung-k'an ed., VIII, 7a —7b). This will bring the date of its construction back to the first part of the tenth century B. C. according to the traditional chronology.
121 For more information on the Hung Kou canal system, especially on the courses of the rivers which constituted the system, see Ku Tsu-yü, op. cit. 6a —6b.
122 Shih-chi, XXIX, 2a.
123 This period is called the period of the hegemony of the state of Liang (Wei) by Ch'ien Mu, op. cit. (see above, n. 82), tables, p. 90—91.
124 Tso-chuan LVIII, 9a.
125 For more information on the Han Canal see also Ku Tsu-yü, op. cit., 8a —8b.
126 Between the Chi River and Tzǔ. River on which Lin-tzǔ was situated there was a canal to connect the two. See Shih-chi, XXIX, 2a.
127 The term wu-tao image appears in the Shih-chi several times always in connection with the struggle for the region in which T'ao was located. Chüng Hsüan (127 —200) interprets it as meaning "cross-road." While many a scholar holds it as the name of a place with a definite location, Müng Wün-t'ung image understands it figuratively. He contends that there is no place or road of this name; it refers to the region which was so important to communication in the empire that it was given the name of "cross-road." ("Lun ku shui-tao yü chiao-t'ung," (On the water ways and communication in ancient times), Yü-kung II (1935), No. 3, p. 4.
128 In Chan-kuo ts'ü (Writings on the Warring States), Ssǔ-pu pei-yao ed., XIII, 3a, T'ao and Wei are mentioned together as places of wealth. The text is also quoted in Shih-chi, LXXXIII, 8a. Some commentators regard T'ao and Wei as referring to two persons. They are the Old Gentlemen Chu of T'ao and Prince Ching of Wei to Yen Tu image or the lord of T'ao (Wei Jan) image and the lord of Shang, whose name was Wei, to Wang Shao image In either case, it makes no difference whether "T'ao" refers to the Old Gentleman Chu or the Wei Jan, the noble who owned T'ao as his fief, because both unequivocally refer to the same T'ao and both indicate that T'ao was a place of great wealth. With regards to "Wei" the comments advanced are not convincing. Prince Ching of Wei was not known for his riches; nor was the lord of Shang. Furthermore, if it were the Lord of Shang who is referred to as "Wei," why is he not directly referred to as "Shang," but as "Wei?" as in the case of the lord of 'T'ao? If we understand "T'ao and Wei" as two places equally renowned for their wealth, then we will have no trouble in interpreting the text and the context in which the phrase stands. In the Han-fei-tzǔ "T'ao and Wei" are mentioned as places of great importance. (V, 11a).
129 Shih-chi, CXXIX, 4b.
130 "Warehousing and Trapezite Banking in Antiquity," Journal of Economic and Business History, III (1930—1), 30—31.



In his brief treatment of money in Chinese antiquity, Ssǔ-ma Ch'ien (145—86 B. C.?) mentions tortoise shells and cowrie shells as objects which had been used as currency and indicates that pearls, jade and tin had also been so used besides gold, silver, and the spade and knife coins of bronze.1 Scattered statements in the literature of the Chou period tend to lend credit to his report. However, it is rather unlikely that all of these were used without preference either throughout the whole Chou period, or in all parts of ancient China. This would be true especially after trade had developed to a considerable extent, say from the Ch'un-ch'iu period on, when portability and homogeneity would be required of a currency. Tortoise shells were of limited use in divination only and are too bulky to be conveniently transported. Pearls and jade did not exist in any appreciable quantity. Neither gold, nor silver, nor tin was produced within the borders of the Shang and the Chou kingdoms, and therefore they lacked another requirement of a currency — general availability. Because of their limited availability and consequently high value, they could not have been suitable for use in ordinary business transactions. Among the monetary commodities mentioned by Ssǔ-ma Ch'ien, only cowrie shells, copper and bronze met both these qualifications of a satisfactory exchange medium, portability and availability.

To be sure, neither copper nor cowries were of local origin, but they seemed to have been available in such quantities that they could have served as currency.2 Both were employed in the beginning as barter units, copper in the form of a spade or a knife and cowries as ornaments. Spades and knives were commodities of general use, and cowries owed their desirability as money to their highly ornamental value and their magic functions. This agrees with the general principles on the origin of money laid down by William Ridgeway.

The importation and use of cowrie shells in China has had a long history. It is well known that a specimen was discovered by J. G. Andersson at the neolithic site at Yang-shao-ts'un, Honan. In the graves of Chu Chia Chai, which also contained pottery of the "Yang-shao" stage, a piece of bone was found carved in imitation of a cowrie shell.3 Seventy-three genuine shells were dug out of the old tombs at Tou-chi-t'ai in the county of Pao-chi, western Shensi. They evidently served as both ornaments and for magical purposes.4 The archaeologist Su Ping-ch'i has dated these tombs as belonging to the period of "bent-feet li" (the li is a tripod), which in his chronology is intermediate between the black pottery (Lung-shan type) culture and the end of the Shang dynasty.

When the site of Yin-hsü, at An-yang (northern Honan), which served as a capital to the Shang kings for the last four to five hundred years of their dynasty was excavated by archaeologists of Academia Sinica, large numbers of the shells were discovered. During the first season, in 1928, ninety-six "cowrie shells and large clam shell utensils" were excavated.5 In 1929, a pit designated as Ta-lien yielded "a layer of cowries" together with some bronze objects and stone knives at a depth of 5.6 meters, between a layer of tortoise shells and another of clam shells.6 In 1931, cowries were found in pit E 16 at a depth of 4.5 to 4.9 meters.7 A considerable number – the report says "Very many" – were found in Section B 14 at a depth of 1.1 meters. In 1932, some more were found in Section E 157 at a depth of 2.4 meters together with some pottery, clam shells, stones, tortoise shells and reindeer horns.8 In the same year in the square pit designated as E 181, at the depth of 6 meters, 163 cowrie shells and two large shells were found together with a long lo-shih (melania libertina), tortoise shells, bone plates, stone knives, animal bones, stone ch'ing music instruments, stone vessels, carved stones and clam shells.9 It must be noted that the square pit designated as E 181 and the rectangular pit designated as Ta-lien have been identified by the archaeologists who participated in the excavations as kao or store pits.

Now the question arises: Had the cowrie shells discovered in the Shang remains been used as money? To this question the answers offered are almost unanimously affirmative. Archaeologists such as Li Chi and Tung Tso-pin, who participated in the Yin-hsü excavations, assert that they were.10 So do also Creel and Okutaira and some other Chinese scholars.11 While in an earlier publication Kuo Mo-jo seems to have the idea that cowrie shells were used as money during the Shang time;12 in a later one he expresses the opinion that use of cowries as money began during the transitional period from the Shang to the Chou, 13 which, according to the traditional chronology, is around 1122 B. C. However, none of these scholars has produced any evidence to support their opinions.

Before attempting our own answer to the question, let us first examine the available facts, archaeological as well as historical. In the literature of or concerning the Shang dynasty we have only one statement on the cowries. It is in the chapter P'an-küng in the Shang-shu (Book of Documents), popularly known as the Shu ching. 14 In this document, P'an-küng, the Shang king who moved the Shang capital to Yin (hence Shang is also designated as Yin) on the site of what is now known as Yin-hsü, reproached his ministers for neglecting their duties and coveting "cowries and jade." At the end of the document the king instructs his ministers thus, "You shall not accumulate the huo and the pao and make profits for your own use." "Huo" means "money," and "pao" means "treasures." These words obviously have reference to the "cowries and jade" used in the beginning of the document. Since these objects, or at least one of them, could yield the profit or income indicated in the words of P'an-küng, they obviously had an economic function beyond their ornamental value.

The importance attached to cowrie shells by the Shang and their possible monetary function may also be gathered from certain inscriptions on oracle bones, which are of a divinatory character, and those on bronze vessels, which serve a commemorative purpose. One oracle bone inscription says:

(Divined) on küng-hsü day, (a personal name) declaring the query: Grant (one) p'üng of cowries to the mothers (or wives).15

Another, which is very fragmentary, reads:

Mother (or wife) ten p'üng (of cowries) ( ) the sons.16

These are divinations for granting cowries. The p'üng is the unit which will be discussed on another occasion.

A third oracle bone inscription is a divination asking whether cowries would be captured. It reads:

Divined on wu-shün day, (a personal name) declaring the query: Would there be captured cowries?17

Besides these there are four more fragmentary inscriptions recording the "taking" (or "receiving") of cowries:

  • ... takes cowries.18
  • ... takes one hundred cowries
  • ... takes six hundred cowries.19
  • ... takes cowries.20
  • Divined on ting-hai day, Kuang takes two p'üng of cowries, in the first month. Took.21

The meaning of the character ch'ü (to take) in the inscriptions above quoted is not clear. It may have been used in the sense of "to receive" as Dr. R. S. Britton has indicated. In a number of the Chou bronze inscriptions there are found expressions such as "ch'ü (an undecipherable monetary appellation) five lieh (a weight or monetary unit)," "ch'ü ( ) twenty lieh," and "ch'ü ( ) thirty lieh." Those who take or receive the money (ch'ü) are all ministers of the Chou court. Kuo Mo-jo regards the expression as denoting that a given minister takes (or receives) a given amount of money as monthly salary.22 This interpretation is plausible. However, whether the same character is used in the same sense as it is in the oracle bone inscriptions is hard to say.

According to Tung Tso-pin, inscription No. 1 belongs to the period ending 1281 B. C.; No. 2 belongs to the period from 1240 to 1227 B. C.; Nos. 3 and 4 belong to the period from 1209 to 1112 B. C., which corresponds to the reigns of the last two kings of the Shang dynasty.23

In the inscriptions of the Shang bronzes we more frequently find references to grants of cowries to those who had the bronzes made. There are nine such inscriptions in the Yin wün ts'un (A Collection of the Yin or Shang bronze inscriptions) and the Hsü Yin wün ts'un (a supplement to the former work).24 A typical one reads:

On the day ting-mao the king ordered Tsu-tzǔ to meet (the chief or representative of) the Kuei state at Hsing.Upon his return the king rewarded him with one p'üng of the cowries which were captured in the expedition against Yung. For this he (Tsu-tzǔ) made this tripod in honor of his father I.25

In some of the inscriptions the number of p'üng granted is not recorded, a fact which indicates that the number may be one. The numbers which are given range from one to ten.

The fact that meritorious ministers of the Shang kings were rewarded with cowries which evidently served no ornamental purpose indicates strongly the financial value of the shells.26

In the inscriptions of the Chou bronzes, especially those of the first three centuries of the dynasty, the use of cowries as rewards and gifts by kings and nobles to their inferiors is very conspicuous. Of the 162 inscriptions concerning the royal court of Chou, which have been collected, dated and discussed by Kuo Mo-jo in his Liang Chou chin-wün-tz'ǔ ta-hsi k'ao-shih, twenty-one contain statements of grants of cowries as rewards or gifts. The bronze inscriptions on the vessels classified by him as belonging to the various feudatories are not included in this count.27 The following are two examples in the simplest form:

In the thirteenth month, on the day hsin-mao, the King was at Han. He granted Ch'ien a fief called (the name of the fief, undecipherable). He granted [him also] five p'üng of cowries. In gratitude for the king's favor, he (Ch'ien) had this Chi precious vessel made.28 When King Ch'Sng offered the great Pün sacrifice at Tsung-Chou (capital of Chou ), he rewarded Marquis Hsien, Hsiao, cowries. For this he (Marquis Hsien) had this Marquis Ting vessel made. Pien-yüan [clan].29

Although the reasons for these grants are not expressed, it is obvious that they were made in return for some meritorious service by the beneficiaries. Another inscription on a kuei vessel made by Su indicates that he received cowries and was exempted from further military service on account of his valor in a campaign against the Eastern Barbarians.30 A tripod made by Lu bears a record that ten p'üng of cowries had been bestowed upon him for his contributions in a campaign against a barbarian people in revolt.31

As to the question whether cowries were used as money during the Chou time, we have not the slightest doubt. The well informed great historian Ssǔ-ma Ch'ien says that they were,32 as do the learned scholar Hsü Shün (first century A. D.)33 and the authoritative commentator on the Confucian classics, Chüng Hsüan (127—200).34 In the original text of the I or the I ching (a book for divination) the loss of cowries is expressed as the loss of property.35 In a love ode in the Shih or the Shih ching (a book of odes) a girl sings that after she met her man he gave her one hundred p'üng as a gift.36 As has been correctly commented by Chüng Hsüan, the phrase "one hundred p'üng" means that many p'üng of cowries.37 Since the amount of the gift is very large, it is not likely that the cowries were ornamental objects. In the Li chi (Book of Rules), another Confucian classic, it is advised that if an inferior wants to give gold, jade, or "money cowries" to his lord when the latter is about to go on a trip, it is proper for him to say that he wishes to present some "horse fare" (traveling fare) to his lord's attendants.38 The word "money" in the phrase "money cowries" is huo, which, when used in the Chou literature to denote a gift, may mean gold, jade, silk, furs or any other valuable. Since in this case gold and jade, which are also huo for gift, have been specifically mentioned, the character in the present text can only mean "money" modifying "cowries."

The use of cowrie shells as money is also corroborated by archaeological finds. In recent years, large numbers of cowries have been discovered in the graves of the Chou period. The conditions in which they were discovered point to the fact that they were money. In a tomb uncovered in 1923 at Hsin-chüng in central Honan, 317 cowrie shells were found.39 The excavation was not performed by trained archaeologists, and therefore exact information is not given; but according to Kuan Pao-ch'ien, these cowrie shells were found in one tomb. They were not deposited at the place where the coffin lay, but in and around a large bronze tripod and a square bronze tsüng vessel at the southern end of the grave.40 Judging from the large number of bronze vessels found, the tomb must have contained the bones of a member of the high nobility. Since cowries occurred in such quantity to preclude their use as magical objects, and because of their disposition at a distance from the body, Kuan Pao-ch'ien was led to conclude, and rightly so, that these cowries functioned as mortuary gifts, money to be used by the dead.41

From the spring of 1932 to December, 1933, about a hundred tombs of the Chou period were found in Chun County in northern Honan, about forty miles southeast of Yin-hsü. From some of them 3472 cowrie shells were recovered. These cowries, "mostly" deposited around the burial vessels, were strung together.42 According to Kuo Pao-chün, the archaeologist, these were tombs of members of the ruling house of Wei, ranging in date from its first ancestor, K'ang-shu, who was enfeoffed in 1113 B. C. (traditional date), to Duke Ch'üng (634—589 B. C.).43 The earliest datable bronze vessel, a tsun, is said to have been made by Duke Wu (812—758 B. C.).44 Since their disposition is like that of the shells in the Hsin-Chüng tomb (namely, in or around the burial vessels at a distance from the corpse), we are inclined to regard them also as money, and the money for the dead is an imitation of that for the living.

If all this evidence should fail to convince our readers that cowries were used as money during the Chou period, we offer one more piece which should be sufficient to rid them of any doubt. This is an inscription on a bronze tsun vessel. It reads:

Lord of Chü, Yüan, had this precious vessel made. He used fourteen p'üng of cowries.45

The inscription is probably of early Chou origin.46 No record can be more definite. Since cowrie shells were used as money by the Chou people and since such usage is recorded on the above cited vessel, a natural inference is that the cowries mentioned as rewards and gifts on other bronze vessels may also be considered as money.

So much for the monetary nature of cowrie shells used by the Chou people as recorded in their bronze inscriptions. Are the cowrie shell grants recorded in the inscriptions of the Shang dynasty of the same nature? Our answer is affirmative. These are the reasons. First, a comparative study of the bronze inscriptions of both the Shang and the Chou periods in which the grants of cowrie shells are recorded and of which examples have been quoted above reveals an unmistakable similarity between the two in the way the cowries were granted, both in the formulation of the statements of the grants and in their economic implications. Since the cowrie shells in the Chou inscriptions are money, those recorded in the Shang inscriptions of similar nature must be the same.

Secondly, we know that King Wu, conqueror of Shang and founder of Chou, already made grants of money, as evidenced by the inscription on a chih wine vessel which records the courtier Tan as the recipient of such a gift.47 In view of the fact that the Chou people were rather backward economically, it does not seem likely that the founder of this dynasty should have initiated such an advanced institution as money. Therefore, the use of cowrie shells as money must have originated in the Shang period.

As we have stated before, the Shang people had already engaged in trade and carried their business to distant regions. Under those circumstances, their use of money was not only natural but probably inevitable. Thus we arrive at the conclusion that the Shang people used money (of which cowrie shells were one form), and that the cowrie shells discovered in the Shang remains must have been money or material to be used as money.48

This conclusion does not, of course, exclude the possibility that the Shang people also used cowries for other purposes. It is almost certain that they served also as ornamental and magical objects, and it is as such that they acquired value, which led to their use as money. Most likely their function as ornamental and magical objects continued long after other more satisfactory media replaced them as currency.

End Notes

1 Shih-chi, XXX, 20b —21a. The historian's terms for gold, silver and bronze are "yellow," "white" and "red" metals. So far as we know, gold and silver were never cast into any form of coin; they were used in the shape of bullion. Possibly so was bronze for some time.
2 The existence of large quantities of copper during the Shang and the Chou periods can be gauged from the large amount of the bronze wares, bronze weapons and other bronze objects which are preserved and discovered today.
3 J. G. Andersson, Children of the Yellow Earth, 1934, 323.
4 Su Ping-ch'i image Tou-chi-t'ai kou-tung-ch'ü mu-tsang, 1948, 17–71, 233—4
5 Tung Tso-pin, "Chung-hua-min-kuo shih-ch'i-nien shih-chüeh An-yang Hsiao-t'un pao-kao-shu," An-yang fa-chüeh pao-kao image, I (1929), 35.
6 Li Chi image "Min-kuo shih-pa-nien ch'iu-chi fa-chüeh Yin-hsü chih ching-kuo chi ch'i chung-yao fa-hsien," An-yang fa-chüeh pao-kao, II (1930), 236.
7 Li Chi, "An-yang tsui-chin fa-chüeh pao-kao chi liu-tz'ǔ kung-tso chih tsung-ku-chi," An-yang fa-chüeh pao-kao, IV (1933), 565.
8 Shih Chang-ju image "Ti-ch'i-tz'ǔ Yin-hsü fa-chüeh: E ch'ü kung-tso pao-kao," An-yang fa-chüeh pao-kao, IV (1933), 719–720.
9 Shih Chang-ju, op. cit., 723.
10 In one of his articles on the findings at Yin-hsü, Li Chi says "Both the cowrie and clam shells were carved into ornaments; they were also money in circulation at the time. The salt-water shells were mostly used for money, and the fresh-water shells were mostly used for ornaments." An-yang fa-chüeh pao-kao, IV (1933), 375. This statement is shrouded in ambiguity. It is regretted that Prof. Li, like all others dealing with the problem, fails to offer any evidence to prove his point. Prof. Tung Tso-pin writes, "During the Yin (Shang) dynasty cowries were definitely the important money. Of those discovered [at Yin-hsü] which have a hole for stringing all belong to it (money)." T'ien-yeh k'ao-ku pao-kao image I (1936), 126. He, too, fails to supply his reasons.
11 Creel, Birth of China , 69, and 91 —2. Okutaira, Tōa senshi, II, 24a —24b. Lo Chün-yü is one of these Chinese scholars. His opinion is found at the end of his work Yin-hsü ku ch'i-wu fan t'u-lu. The scholars quoted in this and the preceding notes are those who have expressed their opinions directly in connection with the Yin-hsü finds. Not in direct connection with the Yin-hsü finds, but rather in a general manner, some other scholars have expressed similar opinions. Terrien de Lacouperie has alleged that cowrie shells had been used as currency as far back as the twenty-third century B. C. ("Metallic Cowries of Ancient China," Journal of the Royal Asiatic Society of Great Britain and Ireland, XX, 1888, 428 ft.). Henry A. Ramsden says that cowrie shells were used as money during the Shang dynasty ("The Cowry Currency of Ancient China," The Numismatic and Philatelic Journal of Japan, II, 1913, 161). A number of other Western numismatists share their opinion. Henry E. Gibson even goes as far as to allege, "It has been definitely determined that during the Shang period (1766—1122 B. C.) the use of Cypraea shells as a medium of exchange was well established and actually the money of the period." ("The Use of Cowries as Money During the Shang and Chou Periods," Jour. N. C. B. Royal Asiat. Soc. LXXI, 1940,33.) But if we should ask how and when the question "has been definitely determined," Mr.Gibson may not be able to answer. Copied from one another a belief has become a conviction, and a conviction a fact.
12 Ku-tai shü-hui yen-chiu (a study of ancient Chinese society), 1930, 3rd print, 251.
13 Pu-tz'ǔ t'ung-tsuan, 1934, III, 101.
14 The P'an-küng chapter in the Shang-shu has been accepted as a reliable document concerning the Shang dynasty by such leading Chinese historians as Wang Kuo-wei in the past (Ku shih hsin chüng), and Kuo Mo-jo at the present (Shih p'i-p'an shu, 1945, 15). But Creel contends that it is a forgery, and certainly it cannot be regarded as reliable historical literature. The reasons which Creel lists to prove the document a forgery are: 1) During the Shang time its capital was called Shang or ta i Shang, "the great city Shang," while in the document it is called Yin; 2) In the inscriptions on the oracle bones the chief Shang deity is the Shang Ti (god on high) while in the document it is T'ien (Heaven); 3) The style of the writing does not seem to be of Shang origin because it is smoother than the inscriptions or the authentic books of the Western Chou period; 4) The document fails to state the reasons why P'an-keng moved his capital to Yin, a fact which shows the forger to be ignorant of them. (Studies in Early Chinese Culture, Baltimore, 1937, 65—67)-Creel's accusations show some legitimate ground to doubt the writing as a contemporary original document, but they are not sufficiently strong to prove it a forgery. The factors which led P'an-keng to move his capital to Yin may have been known to his people, and thus a statement on that point may have not been necessary. The question of the style, while plausible in one respect, is uncertain in others. Any comparison of the writing with the terse inscriptions on the bronze vessels for obvious reasons is inappropriate. Compared with Chou literary documents its smoothness is very slight, and it is a matter of degree not of substance. I, for one, cannot be sure if I understand correctly more than half of the document. Other scholars may not be able to claim much more. Furthermore, in intellectual training and in cultural life as a whole the Shang people were much superior to the Chou. It is not unnatural to find better writing by the hands of Shang intellectuals. Later revisions and copyist changes may have taken place and altered the original composition somewhat. But revision does not imply forgery. Creel's discussion on the god of T'ien is based on negative evidence. The Chou people worshipped both T'ien and Shang Ti. The Shang people may have done the same. The phrase which Creel reads ta i Shang is also written t'ien i Shang in no less than three oracle bone inscriptions. On the name of the capital of P'an-keng, Creel shows that he has not studied the inscriptions on the oracle bones carefully. In those inscriptions "Shang" or the "ta i Shang" or 't'ien i Shang" does not refer to the capital in Yin to which P'an-küng moved but to the older capital of Shang which was what is now called Shang-ch'iu in eastern Honan (See Tung Tso-pin, Yin li p'u, 1945, Part II, IX, 62f). Shang-ch'iu was also the capital of the Sung state during the Chou dynasty, and the ruling house of the Sung state were the descendents of the Shang kings.
15 Lo Chün-yü, Yin-hsü shu-ch'i hou-pien (A supplementary collection of the inscriptions of the oracle bones found in Yin-hsü), Part II, 8, no. 5.
16 Quoted in Okutaira, Toa senshi, II, 25a.
17 Lo Chün-yü, Yin-hsü shu-ch'i ch'ien-pien (An earlier collection of the inscriptions of the oracle bones found in Yin-hsü), V, 10, no. 4.
18 Quoted by Tung Tso-pin in his "An-yang Hou-chia-chuang ch'u-t'u chih chia-ku wün-tzǔ" (The oracle bone inscriptions unearthed at Hou-chia-chuang village in An-yang," T'ien-yeh k'ao-ku pao-kao (Journal of Field Archaeology), I (1936), 126.
19 Ibid.
20 Ibid.
21 Ibid. The translation of this inscription is also found in Britton, Fifty Shang Inscriptions, 1940, 15.
22 Liang Chou chin-wün-tz' ǔ ta-hsi k'ao-shih, I, 57b.
23 Tung Tso-pin, ibid., and Yin li-p'u, Part I, I, 2a.
24 The second work may contain some inscriptions belonging to the early Chou, but that would not affect our case.
25 Lo Chün-yü, Yin wün ts'un (A collection of Yin inscriptions), I, 8b, no. 2. Transcriptions of the inscription are also found in Yü Hsing-wu image Shuang-chien-ch'ih chi-chin-wün hsüan, III, Part 1, 4a —4b; and Wu K'ai-sheng image Chi-chin wün lu, I, 10b. Yü Hsing-wu, like some other epigraphers, reads the character for "Kuei" as hsi or "western." Wu K'ai-shüng reads the character for "Hsing" as hsiang image The phrase "which were captured in the expedition against Yung" is given by Wu.
26 There is no literary or archaeological evidence which indicates that the male nobility of the Shang or the Chou dynasty used cowries as personal decorations.
27 Creel reports that he found thirty-three inscriptions of the Chou period in which "cowries are said to have given, as reward for service or as a mark of esteem, to vassals by their superiors." (Birth of China , 92). This number should include the twenty-one inscriptions which have been dated more closely by Kuo Mo-jo and which we have used in our present study. Creel's number should also include the inscriptions on the bronzes belonging to the Chou feudatories. Jung Küng lists image fifty-one inscriptions which have the character pei (cowries) (Chin-wün pien, 1939, VI, 13b—14a). In the bronze inscriptions whenever cowries are mentioned they are always in the form of a grant. Jung Küng's fifty-one include a few inscriptions of the Shang period.
28 Kuo Mo-jo, Liang Chou chin-wün-tz'ǔ ta-hsi k'ao-shih, I, 15b.
29 Kuo Mo-jo, op. cit., 31b.
30 Kuo Mo-jo, op. cit., 23a—23b.
31 Kuo Mo-jo, op. cit., 27a.
32 Shih-chi, XXX, 21a.
33 Shuo-wün chieh-tzǔ under the character pei image or cowrie.
34 Shih ching, X, 5a, note.
35 I ching, under the Hsün image divinational diagram.
36 Shih ching, X, 5a.
37 Ibid, note.
38 Li-chi chu-shu 1871, XXXV, 3a.
39 Chin Yün-o image Hsin-chüng ch'u-t'u ku-ck'i t'u-chih hsü-pien (A supplementary illustrated catalogue of the ancient objects unearthed in Hsin-chüng), 1923, 8b.
40 Kuan Pao-ch'ien image Hsin-chüng ku-ch'I t'u-k'ao (Studies with illustrations of the ancient objects unearthed in Hsin-chüng), 1940, XI, 20a—20b.
41 Ibid.
42 Kuo Pao-chün image "Chün Hsien Hsin-ts'un ts'an-mu chih ch'ing-li," T'ien-yeh k'ao-ku pao-kao, I (1936), 193—4.
43 Kuo Pao-chün, op. cit., 200.
44 Sun Hai-po image Chün-hsien i-ch'i (The bronze vessels recovered in Chün County), 1937, 13a—14b.
45 Wu Shih-fün image (1796—1856), Chün-ku-lu chin-wün (A collection of the ancient inscriptions: the bronze inscriptions), 1895, XXII, 3a. While this scholar reads the combined characters as "fourteen," other epigraphers read it as "thirteen." Kuo Mo-jo regards "Chü Po" (Lord of Chü) as the tz ǔ (style) of the person who made the vessel. He reads "Yüan" as "Huan," which, he thinks, is the name of the person. (Chin-wün ts'ung-k'ao, 1933, 106b—107a).
46 Wu K'ai-shüng holds this opinion. Chi-chin wün lu, II, 20a.
47 Kuo Mo-jo, Liang Chou chin-wün-tz'ǔ ta-hsi k'ao-shih, 2b.
48 The cowrie shells discovered in the Shang remains are said to be of two types(?). One type has its dorsal side ground flat while the other has a hole instead. It is not known whether the grinding of their dorsal side has any relation with their use as money or ornamental objects, though Prof. Tung Tso-pin has voiced the opinion that those with their dorsal sides ground flat were used as money. All the cowrie shells recovered from the Chou tombs seem to have been ground flat.


Although cowrie shells have been discovered by the thousands in Shang remains and Chou tombs, detailed descriptions of them are regrettingly lacking. Some of the shells discovered between 1928 and 1932 in Yin-hsü are described by Li Chi, the archaeologist, as "saltwater-shells."49 Tung Tso-pin, who also took part in the excavations, has informed the author that some of the cowrie shells found in the Yin or Shang ruins have their dorsal side ground flat and that others have been holed only. The Japanese scholar Nishimura Shinji reports that he possesses a cowrie shell which was originally deposited in a bronze lei vessel discovered in Yin-hsü. He identifies his shell as of the species Cypraea moneta.50 It is 22 mm. in length and 17 mm. in width. Each of the outer labia on the front has twelve horizontal nicks. The color of the outside surface seems to have been a lustrous light yellow. However, most of the enamel part has been worn away. It has a large hole on its dorsal side, which, as he says, is probably due to the shell having been ground down to a flattish oval shape.51

The collection of the Museum of The American Numismatic Society has twenty-four specimens of cowrie shells (cf. Plate I, 1—3). Of them twenty are from the collection of Henry A. Ramsden, and four are from that of John Reilly, Jr. They are of the species of Cypraea moneta and Cypraea annulus. The Cypraea moneta has a purple top on the dorsal side, and the Cypraea annulus has a yellow ring around the top. The species of those whose dorsal sides have been ground flat or broken off can be identified by a comparison of their ventral sides. Three of the four specimens in the collection of Mr. Reilly have been badly decomposed, their enamel having disappeared altogether. Their dorsal sides are ground flat. The inside of the shells is filled with aged earth, a sign of long burial under ground. In size they average 20 mm. in length and 15 mm. in width.

According to Ramsden his twenty specimens formed part of a find discovered in the neighborhood of Chang-tü-fu, Honan, May, 1913.52 Chang-tü-fu is the modern An-yang, where in Yin-hsü, or the "ruins of Yin (Shang)," the bronzes and oracle bones of that dynasty were dug up. If the report is correct, Ramsden's specimens may have come from the same ancient site. These specimens (cf. Plate I, 1, 3) have been described and illustrated in Ramsden's article, "The Cowry Currency of Ancient China," published in the Numismatic and Philatelic Journal of Japan in 1913. They are of various sizes. The largest measures 30.5 mm. in length, with thirteen horizontal nicks on the right and twelve on the left of the ventral side. The smallest measures 12 mm. in length with eleven horizontal nicks on the right and twelve on the left of its ventral side. Six of them have their dorsal side ground flat. Fourteen others have one or two apertures. The holes on three are so large that it appears as though the shells had been ground down to that point. All are discolored, and the surfaces of most of them are decomposed.

As is generally known, both Cypraea moneta and Cypraea annulus grow in large numbers in the Indian Ocean and some parts of the South Seas. Neither species is found in the China seas. In a letter to the author, Dr. William Ingram, an expert on mollusks, says that the places nearest to China where cowrie shells are reported to have been found are the Japan seas in the Japanese Islands, Formosa, the Hong Kong area, the Cochin coast, the Philippines, and certain areas on the Malay Peninsula. He adds that they are also reported to have been found on the China coast, but this is not confirmed. Dr.John C. Armstrong of the American Museum of Natural History mentioned to the author the area around the Tizard Banks, the Ryukyus and Hakodate in Japan as the habitat of the cowries which is nearest to China, observing that he was skeptical about reports on other places because they were not made by scientists. Even if reports on these doubtful localities were reliable, the number of cowrie shells growing at them cannot have been large.

Some of the cowrie shells found in the Shang and the Chou remains may have come from any of the places mentioned above. Some may have been brought from as far as the coast of the Indian Ocean and the islands off the Indian Peninsula. A passage in the later edition of the Bamboo Annals (Chin-pün chu-shu-chi-nien) states that in the first year of King Li of Chou (877—842 B. C. in the traditional chronology) the state of Ch'u in the present Yangtze valley sent cowries to the Chou court as tribute. Although the passage is found only in the later edition, there is reason to believe that the record is reliable.53 Similar payments of which we have no record must have been made in other years.

Next to tribute, war booty was probably the main source of cowries for the court of Chou. A number of bronze vessel inscriptions refer to "capture of cowries,"54 an event which must have been of some importance to be commemorated by the casting of a bronze vessel. While there is no mention of the exact amounts of such loot, it may be assumed that it was considerable in quantity.

Besides these sources of cowries a third can be reasonably assumed, ordinary exchange with the people who possessed, or had access to, them and the amount so acquired by the Chinese may have been considerable.

End Notes

49 An-yang fa-chüeh pao-kao, IV (1933)> 375.
50 Shinji Nishimura, "Ancient Chinese Coinage and Its Origin," Canton, I (1939)> No. 4, p.26.
51 Ibid.
52 "The Cowry Currency of Ancient China," The Numismatic and Philatelic Journal of Japan, II (1913), 163.
53 In the inscription on a p'an vessel made by Hsi-chia (Liang Chou chin-wün-tz'ǔ ta-hsi k'ao-shih, 143b) and another on a kuei vessel made by Shih Yüan (op. cit.), the Chou king asserts that natives in the Huai River region who were his subjects offered him image The character, made up of the character po image for "white" and pei image for "cowrie," is understood by Kuo Mo-jo as the original form of the character pu image (spade coin or cloth) (op. cit., 144a and 148b). But this suggestion is untenable. Since the character is composed of the character for "white" and the character for "cowrie," there is reason to understand it as signifying "white cowrie." The character might have been coined to differentiate white cowries from cowries (or shells other than cowries) of variegated colors. Both Cypraea tnoneta and Cypraea annulus which have been discovered in the remains of the Shang and the Chou periods are predominantly white. If our suggestion is satisfactory, then "White cowries" must have been a specified tribute offered by the Huai people to the Chou court.
In the inscription on the kuei vessel made by the head of the Huai state (op. cit., 147a) it is stated that the chief of a certain conquered state came to the Chou king to offer image Composed of po image and pei image ( (cowrie) the character has been correctly identified by Kuo Mo-jo and other epigraphers with the character for "white cowries" discussed above (op. cit., 144a). It is interesting to note that the state which offered image as tribute was also situated in the south (in the middle Yangtze River valley). It was from the south that cowrie shells reached the Chou people.
The practice of exacting cowries from the southern states as tribute seems to have lasted for a long time. As late as 179 B. C. the Southern Yüeh state in present Kuang-tung province is recorded to have offered them as tribute to the imperial court of Han (Han shu, XCV, 28b).


Various materials have been used in making imitations of cowrie shells. Lo Chün-yü has reported the discovery of bones carved in the form of cowries in Tz'ǔ-Chou (present day Tz'u County), southern Hopeh, not far from An-yang.55 According to Hamada Kosaku, Lo had in his possession forty such specimens found in Hsin-an County, western Honan, and in T'üng County, southern Shantung, twenty from each place.56 The American Numismatic Society has thirty cowrie imitations in its collection, some of which, according to Ramsden, belong to finds made at Tsinan, capital of Shantung province, and Chang-tü-fu, the present-day An-yang.57 Ramsden gives Ying-yang in central Honan below the Yellow River as another source for some of the pseudo-cowries in this collection.58 Nishimura Shinji and Okutaira Masahiro also report the discovery of such bone carvings at Yin-hsü.59 Both the Museum of Far Eastern Antiquities in Stockholm and Mr. H. E. Gibson possess collections of imitation cowries whose provenance is unknown.60

According to Lo Chün-yü, a stone substitute of a cowrie shell has been found in Yin-hsü.61 A number of stone cowrie imitations are said by Chüng Chia-hsiang62 to have been found in Ho-chien, central Hopeh, and also probably K'ai-füng, eastern Honan. Gibson reports that he had a specimen made of "black stone."63 In the collection of The American Numismatic Society there are three fine stone specimens.

A substitute made of another kind of shell, probably from fresh water, is said to have been discovered in Yin-hsü, An-yang.63a The existence of cowrie imitations of such material is reported by Nishimura and Gibson.64 Yin-hsü is also said to have yielded imitations made of "white marble,"65 which may be identical with some of the stone imitations reported by other numismatists.

Cowrie imitations in bronze are comparatively rare; The American Numismatic Society has six specimens and the Museum of Far Eastern Antiquities has a few.66 They are reported to have been found at Yin-hsü (An-yang, Honan), Tz'ǔ-Chou (Southern Hopeh), Chüng-Chou (southern Honan), and K'ai-füng (central Honan).67

Chüng Chia-hsiang mentions jade imitations on which he gives no particulars.68 Nishimura mentions imitations of "semi-precious stone," clay and iron, apparently excavated in Yin-hsü.69 It seems miraculous, of course, that such small objects of iron can have survived three thousand years in the soil. Gibson reports that he possessed imitations made of "a sort of quartz" and of ivory,70 but does not state their provenance. Articles by Nishimura and Gibson are illustrated with imitations of "mother-of-pearl," "quartz," "semi-precious stones," "ivory," and "shell."

Of all the reported materials from which imitations of cowries have been made, it is only those of bone, stone, and bronze for which we can supply exact descriptions. The thirty bone imitations at the American Numismatic Society (cf. Plate I, 4—7), originally in the Ramsden Collection, were described by H. A. Ramsden in his article "Cowrie Substitutes Used as Currency in Ancient China."70a These may be divided roughly into three groups as follows:

1. Generally large and thick, the largest being 26 mm. in length. Somewhat lozenge-shaped. Convex surface. Roughly carved. One hole, either at center or close to one end. Horizontal nicks along ventral groove (sixteen being the largest number) reaching as far as edge of surface. Reverse side, unpolished. Color, dark green.2. Average 19 mm. in length. Two holes symmetrically placed on reverse side. No horizontal nicks along ventral groove. Reverse, flat and polished. Colors, yellow and greenish gray.3. Size about same as second group, largest being 22.5 mm. and smallest 17 mm. in length, but better carved and finished, and more realistic in imitation of a real shell. Oval in shape. Convex surface. With one exception, reverse is flat and polished. Edges along ventral groove curve slightly in resemblance of real shell. Horizontal nicks along ventral groove are short and resemble those of a genuine shell. Two round holes on reverse side. Colors range from yellowish white to brown, and from dark green to chocolate; white ones have patches of fresh green, possibly caused by contact with bronze vessels in which they may have been deposited.

Belonging to the third group of bone imitations are two found in a lei vessel which was unearthed at Yin-hsü according to Nishimura. His detailed observations on them are as follows:

One is nearly symmetrical and forms an oval shape with pointed ends. Either edge of the deep groove in the center is marked with thirteen horizontal nicks. The top side is convex and brown, but the reverse side is flat and smooth and somewhat lighter in color. The holes at either end together with the central groove probably served to string the shell to other things. The other imitation is nearly the same as the first, though a little longer and more slender. Its lower part is narrower than the upper part. The central groove, which is marked with oblique nicks on both edges, is also wider at the upper part. The color is light brown on the reverse side, and the pale greenish tincture at the base of the right side is probably due to a saturation of bronze rust. There are two holes in the back. The latter is on the whole nearer to the real cowrie shell, as it is 23 mm. long, 17 mm. wide and 6 mm. high, while the former is 21 mm., 18 mm. and 5 mm. respectively.71

Certain cowrie imitations discovered in T'üng County in southern Shantung and described by Hamada Kosaku 72 also belong to the third group, characterized as they are by a realistic resemblance to real shells. Other specimens reported by the same scholar as discovered at Hsin-an in western Honan are roughly carved and have only one hole in the back. Because of the hole they may be considered as belonging to the third group of our classification also.

Summarizing the above descriptions we may outline the characteristic features of the three groups of the bone imitations of cowries as follows:

  • 1. Roughly carved, somewhat lozenge-shaped, reverse not worked, one hole, large size.
  • 2. Also roughly made, nearly oval-shaped, no horizontal nicks along the ventral groove, two holes, reverse flat and polished, medium size.
  • 3. Finely and more artistically and realistically carved, polished on both sides, oval-shaped, two holes, smaller than Type One.

This general classification, however, is not intended to convey the impression that specimens within one group are the same in every aspect. As a matter of fact no two specimens have been found exactly alike in every detail. We group them together on the ground that their minor variations are comparatively insignificant as against the main features they share in common.

The three stone imitations at The American Numismatic Society, all of which are finely carved, measure 25 by 17.5 mm. ( Plate I, 8), 22 by 16 mm., and 23 by 17 mm. All are oval in shape with one end slightly narrower than the other as with the bone imitations of the third type. The surfaces are carved and polished, while the reverses are flat. The ventral groove of each runs across the whole surface, with curved edges resembling those on a real shell. The incised nicks along the groove are horizontal or nearly so. There are two holes on the reverse side positioned at the tip of either end. In color, they are gray with a strong green tinge.

Chüng Chia-hsiang has published illustrations of two stone imitations in his possession, which differ greatly in workmanship. One appears to be finely carved, resembling very much the real shell, like the specimens at The American Numismatic Society, while the other is lozenge-shaped, without the central groove, and bears little resemblance to a cowrie shell.73 Thus we find that stone imitations vary considerably in workmanship.

The American Numismatic Society has six bronze imitations, some of which were originally in the collection of H. A. Ramsden and have already been described by that numismatist.74 All six are oval in shape. Excepting on one specimen ( Plate II, 1), one end is slightly narrower than the other (as with the bone and stone imitations). They are made from thin bronze plate, convex on their top side and concave on the bottom. Again with only one exception ( Plate II, 1) all have a central groove which, unlike those on bone and stone imitations, does not reach the ends. One piece ( Plate I, 9) lacks horizontal nicks along the groove, while on two others (cf. Plate I, 10) the nicks are distinct. Two pieces ( Plate II, 2, 3) show gilt on their top surfaces. The sixth specimen differs from the others: It is thicker, lacks the central groove and horizontal nicks, but it has the two holes, one at either end. The six pieces measure 24.5 by 16 mm., 24 by 16 mm., 24 by 15 mm., 28 by 17 mm., 23 by 16 mm., and 25 by 18 mm.

Without a single exception, numismatists and historians who have written about the various imitations of cowries agree that they were used as money in ancient China, as real cowrie shells had been. They believed that the substitutes resulted from a shortage in supply of the real shells. As plausible as this opinion seems, it fails to take into account one very important factor, the wide difference in value of the materials from which the substitutes were made. Further, even among substitutes made of the same material there is a considerable range in quality of workmanship.

Let us consider bone substitutes, as an example. With the sole exception of Ramsden, who calls them "horn substitutes," all numismatists regard them made of bone, but of bone of unknown origin. Mr. A. G. Goodwin of the American Museum of Natural History in New York City excludes the possibility that they are made of horn, because the structure of animal horn is not as hard and fine as that of our bone specimens. He was doubtful they were of a special kind of bone, for to distinguish one kind of bone from another when cut into such small pieces is practically impossible, especially for the average public which would have used them. In the opinion of Dr. Nils Nelson, also on the Natural History Museum staff, the colors of the bone imitations may not be original. More likely, they are the result of chemical reaction set up by contact with other material when buried underground. The green color was obviously produced by contact with bronze vessels.

We are thus faced with a question not raised heretofore. If the material of these imitations is ordinary bone, they could have been carved in great quantity without difficulty, and obviously they could not have been accepted as equal in value to the real imported shells. This would hold true in greater degree for imitations of such an inferior material as stone. It seems more likely that bone and stone substitutes for cowries were used either as burial money, as ornaments, or as magic objects by those who could not afford the real shells.

If substitutes had been used as money, they could not all have been in circulation simultaneously with real shells, for they could not have been regarded as equivalent to them. Adding to bone and stone substitutes those made of mother-of-pearl, jade and quartz, the situation is more complex still. In fact, it is so complex, that it is inconceivable they were all used as money, unless we assume that during the Shang and early Chou period there existed a complicated monetary system in which the real cowrie and each variety of its substitutes had specific comparative values and definite rates of exchange. Such a condition, however, postulates a highly centralized governmental authority with full power both to issue and rigidly control money. Whether the courts of Shang and Chou had such power is a question. It was probably on account of these difficulties that Lo Chün-yü advanced the idea that substitutes were money of different times! According to him, the shell substitutes made probably of mother-of-pearl appeared first, and the substitutes made of bone and copper, the latter being the forerunner of the so-called I pi ch'ien or "Ant nose money," followed successively.75 However, his chronological order does not include the substitutes made of other materials, and he is silent on the point whether the appearance of a new substitute means the replacement of the old. Like other numismatists, Lo Chün-yü also overlooked the problem of the relationship of the various imitations to the real shells, which remained in use throughout the entire period in which imitations were made. Lo's opinion is no more than a hypothesis which may prove untenable in the future. However, one of his points is worth noting, that the copper cowrie substitute is the forerunner of the so-called Ant Nose Money.

End Notes

63a Yin-hsü ku-ch'i-wu t'u-lu, illustration no. 21.
70a The Numismatic and Philatelic Journal of Japan , III (1914), 17.
54 For examples see Kuo Mo-jo, Liang Chou chin-wün-t z' ǔ ta-hsi k'ao-shih, 25a and 28a.
55 Yung-lu jih-cha (Coole 392), photostated., 10a, and Yin-hsü ku-ch'i-wu t'u-lu (An illustrated catalogue of objects discovered in Yin-hsü), 1916, the first of the comments attac hed at the end of the catalogue.
56 "Shina kodai no gaika nitsuite" (Regarding the cowrie money in ancient China), Tōyō Gakuhō image II (1912), 264—273.
57 Ramsden, "Cowry Substitutes Used as Currency in Ancient China," The Numismatic and Philatelic Journal of Japan, III (1914), 17.
58 Ibid.
59 Shinji Nishimura, "Ancient Chinese Coinage and its Origin," Canton, I (1939), 4, p. 26.
60 J. G. Andersson, Children of the Yellow Earth, 1934, 323. Gibson, "The Use of Cowries as Money during the Shang and Chou Periods," Jour. N. C. B. Royal Asiat. Soc., LXXI (1940), 33—45, Plate C.
61 Yung-lu jih-cha, 10a.
62 "Shih-pei erh p'in" (Two specimens of stone imitation of cowries), Ch'üan-pi (The Chinese Numismatics Bi-monthly ), No. 8, 35.
63 Gibson, Ibid.
64 Nishimura, ibid. and Gibson, ibid.
65 Nishimura, op. cit27.
66 Andersson, ibid.
67 For Yin-hsü and Tz'ǔ-Chou see Yung-lu jih-cha, 10a, and Yin-hsü ku-ch'i-wu t'u-lu, first comment attached at the end of the work. For Chüng-Chou and K'ai-füng see Chüng Chia-hsiang, Chüan-pi, No. 8, 35. See also Okutaira, Tōa senshi, II, 77b.
68 Chüng Chia-hsiang, Chüan-pi, No. 8, 35.
69 Nishimura, ibid.
70 Gibson, op. cit., Plate C.
71 Nishimura, op. cit. 26. His statement that one substitute is nearer to the real cowrie shell than the others because it is 23 mm. long is not correct. The size of the real shells varies considerably. In the collection of The American Museum of Natural History the largest specimen of the Cypraea moneta measures 43 mm. long, and the largest of the C. annulus measures 29 mm. long.
72 Hamada, Tōyō Gakuhō, II (1912), 264—273.
73 Chüng Chia-hsiang, Ch'üan-pi, No. 8, 35.
74 Ramsden, The Numismatic and Philetelic Journal of Japan , III (1914), 16.


The I Pi Ch'ien (ant nose money), also known as Kuei-lien ch'ien (ghost face money) or Kuei-t'ou ch'ien (ghost head money), seems to be true metallic coinage.76 Inclusion of it in the present section on non-metallic currencies is warranted only by the fact that this type of coin is the last of the imitations of the cowrie shell.

Ant nose money was never mentioned by Chinese historians until certain specimens discovered during the Sung dynasty came to the notice of numismatists of that time. Its resemblance to the cowrie shell was recognized by Li Tso-hsien77 in 1874, and this point, spread by Lo Chün-yü, became accepted by practically all numismatists. It was Hung Tsun, author of Ch'üan chih (Coin Catalogue), who, in his introductory section, first considered it to be real coinage. We learn from him that the term I pi ch'ien had become a popular designation for this money as early as the twelfth century.

The original meaning of the term, I pi ch'ien, is not known exactly. Ma Ang regards it as signifying "ant and nose money," not as "ant-nose money" as understood by many. It is his opinion that "ant" comes from the pictoral impression of the legend image which looks like an ant and which appears on one group. "Nose" comes from that part of the legend image which looks like a human nose and which appears on another group.78 So far, this is the best interpretation, and may well be true. However, neither of these two terms was the original nomenclature, nor, strictly speaking, are they designations for this type of money.

The term "ghost face money" is not difficult to explain. The legend image mentioned above, appearing on an oval convex bronze piece, has a strong likeness to a caricature of the human face of a kind that often strikes the Chinese as a representation of a being of the other world. The term "ghost head money" can be explained in the same manner.

Since Hung Tsun's time, more varieties have been discovered. Some bear the legend chün image some hsing image some chin image and some image this last being undecipherable. Ch'u Shang-ling quotes Liu Shih-lu, a noted numismatist, as having reported a seventh variety whose legend reads t'ao image 79 of which we have seen neither an actual specimen nor an illustration.

Of the six known varieties, the legends of three are easily recognizable. They are mentioned above as chün, hsing and chin, meaning in modern nomenclature "lord," "go," and "metal" respectively. The decipherment of image as image (shih-huo) meaning "ten huo" is unacceptable, because the character supposed to be huo does not agree with the forms of this character in other coin inscriptions or in the inscriptions on bronzes of the Chou period, when these coins are said to have circulated.

For the legends of image and image various decipherments have been suggested. Numismatists have read the first one as chin image (name of a state), or k'u image (weep), or tang-pan-liang image (equal to half ounce), or hun-tien-shui image (to pacify water), or pei image (cowrie), or chi image (small). For the second legend there are the suggestions of ssǔ image (a grave), tang-kuo-liu-chu image (each equal to six chu), yu-t'u-chih-pün image (the basis for possessing the land), and lo-i-chu image (one chu of the city of Lo). It is not necessary to repeat the reasons for which these decipherments have been made. It seems equally unnecessary to dwell on the reasons why they are unsound. We may state, in short, that none of them are epigraphically or philologically sound. Of all of these readings, chin for image and "one chu of Lo" for image are regarded as the best. Judging from the way the inscriptions on the knife, the spade, and the Yüan-chin coins are made, image seems to be a mint name or a serial character. This lends weight to its decipherment as chin. But this decipherment not only meets difficulties in ancient epigraphy, but also does not agree with the fact that this type of money has never been discovered in the territory of the state with this name. Nor is there a town or city with the name which is nearby places where they have been found.

The reading "one chu of Lo" does seem convincing, but epigraphically such a reading is unpermissible. What is supposed to be the lower component part (k'ou meaning "mouth") in the character lo is actually the outline of the hole of the coin. Even if it were a part of the character, it could not be read as the character for "mouth." "Mouth" in this form is not found in inscriptions on coins or bronzes of the Chou period.

The main features of this type of money have been described by Ramsden,80 whose collection forms part of the collection at The American Numismatic Society. The Society has 121 specimens, all of which are of bronze, though some of them have been so oxidized that they appear to be of some kind of stone. They are oval in shape, their observes being convex and reverses flat. All have a hole at one end, though on some the hole does not go completely through. The tip with the hole is slightly narrower, reminiscent of the physical appearance of the cowrie shell. The weights in grams of fifteen specimens in fairly good preservation are as follows: 3.03, 2.60, 2.62, 3.04,; 2.70 3.15, 3.05, 2.81, 1.78, 1.42, 1.40, 1.98, 4.30, 4.42, 3.72.

The provenance of ant nose money is widely spread. The best known finds are those made at the Ch'i-Ssǔ-li village in Ku-shih County in southeastern Honan.81 Ch'u Shang-ling reports that in 1783 several thousands were unearthed during the digging of a canal in Shih-ku-t'an in Chiang-ning (modern Nanking).82 Liu Yen-t'ing is quoted as saying that ant nose money was also found in Ch'ang-an (modern Sian) in Shensi.83 Ch'in Pao-tsan records that some were found in a bronze chileh vessel unearthed outside of the city wall of SuChou (T'ung-shan being its modem official name) in Northern Kiangsu.84 Ma Ang states that some have been seen in Szechuan.85 Kao Huan-wün reports that thousands of them were discovered during railway construction in Honan.86 Since his book was published in 1908, the railway must have been the Peiping-Hankow line which cuts through the central part of Honan Province in a north-south direction. However, the exact location of the site is not known. The most recent discovery reported was made in 1936 in Shou-Chou, officially Shou County, in central Anhui Province. It consisted of about one hundred and twenty or thirty pieces stuck together in an almost unbreakable mass. A few pieces which were successfully broken off are reported to bear the legend image 87

A study of their provenance reveals an interesting point, that, except for their discovery in Ch'ang-an and Szechuan, the reports of which are based on hearsay and consequently doubtful, ant nose money has been found at places well within the territory of the state of Ch'u of the Chan-kuo period (403—221 B. C.). Ku-shih is situated in the central area of the old state, and Shou-Chou, then called Shou-ch'un, was its capital from 241 to 223 B. C. SuChou is the "eastern" part and Chiang-ning was in the "southern" part of the state. This suggests the possibility that ant nose money may have been a currency of the state of Ch'u.88

The attribution of this money to the state of Ch'u has been suggested previously, though the suggestion was not based on a complete study of its provenance. Chu Füng apparently reports the first discovery of "ant nose money" in Ku-shih, and he seems also to have first advanced the opinion that it was issued by King Chuang (613—591 B. C.) of the Ch'u state.89 A popular belief recorded in the Ku-shih County Gazette says that the money was made by Sun Shu-ao, King Chuang's chief minister.90 Both of these assertions originated from a statement in the Shih-chi to the effect that King Chuang regarded the money in circulation as too "light" and attempted to coin "big ones" to replace the "small (light) ones." This monetary reform aroused resentment among his people. On the advice of Sun Shu-ao, his minister, the king abandoned his reform and allowed the "small ones" to continue to circulate.91 Chu Füng and the numismatists who follow his opinion obviously hold that the "ant nose money" must be the "small" coins referred to in the above story. There is no way, however, to prove that this identification is correct. Even if it were, the money could not have been issued by King Chuang or Sun Shu-ao, because, as the Shih-chi text shows, at the time when King Chuang became the king and Sun Shu-ao his minister, the "small ones" were already in circulation.

The money is not recorded in literary sources. The style of their legends, such as the chin and the chün, suggests the possibility of their being late in date. The discovery of the money together with the round coin of the late Chan-kuo period, as reported by Ch'in Pao-tsan, tends to confirm this possibility. According to Ch'in Pao-tsan92 sometime toward the close of the nineteenth century, a bronze chüeh vessel was unearthed outside of the south city wall of SuChou. A round coin and several specimens of "ant nose money" were found in the vessel. The round coin has a square central hole and no rim on its outer circumference, and in illustration measures 28 mm. in diameter. It bears a legend which reads chung shih-erh chu or "Weight, twelve chu," chu being the smallest unit of weight in the coinage of the time. This coin belongs to the very last type of round coin of the third century B. C. and is the direct forerunner of the Pan-liang coin of Ch'in issued after Ch'in unified China in 221 B. C. "Twelve chu" and "pan-liang" (1 liang equal to 24 chu) were different expressions common at that time for the same weight. Since the "ant nose money" was found together with this round coin, it seems to have been in circulation as late as the late third century B. C. The weight of the "twelve chu coin" is about 11 grams, a weight derived from that of the Pan-liang coin of Ch'in. Judging from the fact that in early coinage money was still largely a commodity rather than a token, and weight was still a criterion for its value, the light-weight "ant nose money" must have been an auxiliary currency to supplement the round coin in business transactions.

To summarize our observations we may say that the "ant nose money" seems to have been a currency of the state of Ch'u. The date of its first appearance is unknown, but it was likely still in circulation in the late third century B. C. In all probability it was an auxiliary metal currency supplementing the Yüan-chin of that state.93 During the Chan-kuo period the territory of Ch'u expanded greatly to the north and northeast, annexing much of the territory where the spade coin, and the round coin which succeeded the spade, were the main currency. The conquest of this territory by Ch'u was, by natural assumption, followed by the introduction of the Ch'u money. Hence the discovery of the "ant nose money" together with non-Ch'u money (i. e.,the round coin) in an originally non-Ch'u territory (conquered by Ch'u in 261 B. C.). In the newly incorporated territory the money must have been used as an auxiliary currency, as it had been in its home domain.

End Notes

75 The first of comments appended in Yin-hsü ku-ch'i-wu t'u-lu.
76 Ever since their discovery in the 12th century these small bronze objects, generally known today as "ant nose money," have been regarded as money by numismatists and historians. Although this assumption lacks confirmation in the literary sources and corroboration by evidence other than the objects themselves, it appears plausible and acceptable. Kuei Fu image (1736—1805), a famous philologist, disagreed. He read the legend on one group of this type of money as hun-tien-shui (discussed below, p. 78), and from this reading he inferred that these objects were buried to pacify water (quotations from him are found in the Encyclopedia of Old Coins, II, 19a—19b). As we have pointed out, his reading of the legend is not acceptable, and therefore his suggestion concerning the function of these objects, inferred from the reading, must be discarded. Reading the legend on another group of this type of money as ssǔ image meaning "a grave" or "to dig a grave," Ch'u Shang-ling arrives at the conclusion that they were used in ancient times to be buried with the dead to ward off ants (Chi chin so-chien lu, Coole, 9, XVI, 6b.). However, both his decipherment of the legend on the objects and his explanation of their function are untenable.
77 Li Tso-hsien, Hsu ch'üan shuo, ed. by Pao K'ang (Coole, 202), 1874, 8b.
78 Huo pu wün-tzǔ k'ao (Coole, 222), 1924, 2nd reprint, IV, 20b.
79 Chi chin so-chien lu, XVI, 6b. Also quoted by Kao Huan-wün image (Encyclopedia of Old Coins, II, 19a) and Lo Chün-yü (Yung-lu jih-cha, 10a).
80 "Ant's Nose Money," The Numismatic and Philatelic Journal of Japan, III (1914), 139—140, 165—166.
81 This is said by a few numismatists to have been recorded in the Ku-shih hsien-chih (Local gazetteer of Ku-shih County). See Encyclopaedia of Old Coins, II, 19a and 19b. Chu Füng image author of the Ku-chin tai-wün hsü-lu (Code 255), has also made a statement to the same effect (Encyclopaedia of Old Coins, II, 18b). We have examined the 1786 and 1942 editions of the gazetteer and failed to find the record. As Chinese local gazetteers vary from one edition to another, it is quite possible that the discovery of the Ant Nose Money in Ku-shih is recorded in other editions and not in the two consulted.
82 Ch'u Shang-ling, Chi chin so-chien lu, XVI, 6a—6b.
83 Hu K'un image Ch'ang-an huo-ku pien (A catalogue of old objects obtained in Ch'ang-an), prefaced 1914, Preface.
84 Ch'in Pao-tsan image I-hsia lu (Coole 226), 1903, III, Part 2, 2a—2b.
85 Ma Ang, Huo pu wün-tzǔ k'ao.
86 Encyclopaedia of Old Coins, II, 19a.
87 Hamada Kosaku image Kōkogaku kenkyū, Tokyo, 1940, 429—433.
88 Even if the reports of the discovery of ant nose money in Ch'ang-an and Szechuan, which were outside of the old state of Ch'u, were reliable, they would not necessarily alter this assumption. The area of Ch'ang-an during the latter part of the Chou period was in the territory of the state of Ch'in, which bordered on Ch'u on its southeast, and had close relations with the latter. Szechuan, called Pa and Shu at that time, before it was conquered by Ch'in in 316 B. C. seems to have been under the domination of Ch'u. In this case, too, close relationship between these two areas can be assumed. Therefore, discovery of money of Ch'u in these two areas is not a matter of impossibility.
89 Chu Füng, Ku-chin tai-wün hsü lu, quoted in Encyclopaedia of Old Coins II, 18b.
90 See Encyclopaedia of Old Coins II, 19a. Terrien de Lacouperie follows this interpretation. See his "The Metallic Cowries of Ancient China," Journal of the Royal Asiatic Society, N. S. XX (1888), 428—439.
91 Shih-chi, CXIX, 1b.
92 Ch'in Pao-tsan, op. cit., III, Part 2, 2a—2b and VIII, Part 1, 1a—2b.
93 For the Yüan-chin coins of Ch'u see Chapter VI.


The unit for measuring quantities of cowrie shells as money was the p'üng image whenever a unit was given. The only exceptions are in the inscriptions on an oracle bone cited above (p. 59) and on a Shang bronze vessel,94 which record grants of "one hundred" and "six hundred" in one case and "two hundred" in another, as well as one Chou bronze inscription in which the cowries are reckoned by the lieh.95 The lieh is a monetary unit which will be discussed later (see pp. 207—211).

In inscriptions on oracle bones, the pictograph of the character p'üng is written image or image. On both Shang and Chou bronzes the latter form prevails with only minor variations in a few cases. The first form of the pictograph resembles a string of cowrie shells bent at the middle; it may well have derived from a necklace or other article of adornment which was made up of cowrie shells. It is presumed that when cowries came to be used as money, the unit in which they were reckoned as a medium of exchange was based on the original form in which they were used as ornaments.

Of the two forms of the character, image is probably the original. The difference between the two must have been caused by the fact that inscriptions on oracle bones are incised. When making an incision, a straight line is much easier to execute than a curved one. Hitherto scholars have held that the character p'üng resembles two strings of cowries, hence the English rendering of the term as "double string" by Western scholars.96 The character can be viewed as composed of two strings, but a more realistic hypothesis is that the "double string" is a single string curved at the middle with its pendent ends equal in length and with each end bearing an equal number of cowries.

In a purely imaginative interpretation, H. E. Gibson traces the origin of the character to the way cowrie shells were supposed to have been carried during the Shang and Chou periods. As reconstructed by him, they were strung on two cords each having ten shells. The two cords of cowries were then attached to the two ends of a stick held by hand in the middle.97 This reconstructed picture of the way cowries were carried appears rather unrealistic; and it is obviously based on a notion derived from the later form of the character p'üng rather than from the original.

Earlier than Gibson,L. Wieger made the statement that "The cowries, current money of old China, are offered strung up, often in great quantities, as much as a man can carry with a pole."98 Hence his rendering of p'eng as "man-load."99 Although H. F. Bowker holds that Wieger's views should not be ignored,100 we feel they are worth little serious consideration.

The use of the p'üng, rather than numbers, to measure cowries, strongly supports the theory that they were used as money during the Shang period. It is entirely unlikely that a set number of pieces should have become stipulated for such an ornament as a necklace. On the other hand, a real need to measure exact numbers of cowries would be present when they were being used as currency.

There are controversial opinions on the number of cowries constituting a p'üng. Commenting on a Chou ode which recounts a gift of one hundred p'üng of cowries, Chüng Hsüan (127—200 A. D.) renders the number as five.101 Wang Kuo-wei rejects this number as inacceptable and on the basis of other literary evidence concludes that one p'üng was made up of ten cowries, five on each string (correctly, on each end of the string). He regards Cheng Hsüan's number of five as being the amount on one end of the string, not the total number on both ends which constitute a complete p'üng. 102 Kuo Mo-jo holds approximately the same opinion. He differs from Wang Kuo-wei only on the number in a p'üng when cowries were originally used as ornaments. At that time, he contends, one p'üng may have been made up of two, three or five. But he is inclined to accept Wang Kuo-wei's number of ten for the p'üng as the unit of cowrie shells after they had been used as money.103 On the whole, Kuo Mo-jo's suggestion is very plausible. The p'üng as a form of the cowrie necklace or the like must be distinguished from the p'üng as the unit of the cowrie money. As Kuo Mo-jo has pointed out, in the former case a definite number in a p'üng is not essential, while in the latter case it is mandatory.

So far the discussion of the p'üng as the monetary unit of cowrie shells has overlooked one point which may be of some importance; that is, the standard size or weight of the ten pieces which made up the unit. Creel is of the opinion that, "These little shells are very much alike and there is every indication that one of them had the same value as another."104 With this assertion he declares, "A string of cowries, containing a fixed number, was a string of cowries."105 However, as we have already noted, the cowries excavated in China now in the possession of The American Numismatic Society vary in size. Those recovered from the Chou tombs in Hsin-Chüng, as reported by Chin Yün-o, are also of different sizes. "The largest are more than one inch (Chinese) long, while the smallest are less than one inch."106 Further, "big cowries" (ta pei) are mentioned in one bronze inscription.107 The American Museum of Natural History in New York City City has in its collection a large number of the Cypraea moneta and C. annulus to which species the money cowries of ancient China belong. The largest of the former species is 43 mm. in length, and of the latter species 29 mm. The smallest specimens in both cases are less than 16 mm. It seems naive to assume that the ancient Chinese had regarded the cowrie shells of different sizes as of equal value. In 9 A. D. when the reformist emperor Wang Mang formulated his monetary system, which was intentionally modelled on the Chou tradition as he understood it, he proclaimed cowrie shells to be money too. His shell money was of four classes according to their size. The larger the size, the larger the value. The first class was 0.48 ch'ih or foot (110 mm.) long and over, the second class was 0.36 (82.8 mm.) and over, the third class was 0.24 (55.2 mm.) and over, and the fourth class was 0.12 (27.6 mm.). Those shorter than 0.06 (13.8 mm.) were not to be taken as money. In terms of the copper cash in circulation at the time, the values of the four classes of the cowrie shells were 216, 50, 30, and 10 respectively.108 How much the WangMang system does reflect the old Chou traditions it is hard to say. But we may safely assume that during the Shang and the Chou periods one p'üng of large cowries must have been worth more than one p'üng of the smaller ones. The value of cowries as money presumably originated in their value as ornamental objects. As such the larger ones certainly are more attractive to the eye, and, if any magical effect was ascribed to them, the larger shells must have been deemed more powerful.

As to the actual values of cowries during the Shang and Chou times we have no information except for the cost of the tsun vessel made by Yüan, Lord of Chü. As noted above, the making of the vessel cost the noble fourteen p'üng of cowries. From the style of the expression of the inscription on the vessel, part of which reads, "Used ten p'üng and four p'üng of cowries," Wu K'ai-shüng regards it as of an early origin, similar to that on the known Shang bronzes. If it had been later in origin, the part of the inscription quoted above would have been written as "Used ten and four p'üng of cowries."109 If his judgment is correct, as we think it probably is, the vessel may be one of late Shang or early Chou.

Unfortunately, the present whereabouts of the vessel is not known. This deprives us of an important source of information from which we might have learned the approximate value of the cowries in Chinese antiquity. In his recent work, The Bronzes of Shang and Chou, Jung Küng lists six tsun for the Shang period and two for early Chou. They vary in size, ornamentation and probably also craftmanship. The largest is 1.21 Chinese feet high while the smallest is only 0.55.110

Although the exact value of cowries cannot be known, a general idea of their worth may be gathered from an examination of the bronze inscriptions in which grants of them are recorded. It is a well-known fact that no grant larger than ten p'üng is recorded in the inscriptions of the preserved Shang bronzes. The situation during the first 120 years of the Chou period is the same. Only from King Mu (trad. 1001—952 B. C.) on, larger grants appear. The smallness of the amount of the grants is an indication of the rarity and dearness of cowrie shells at the time. In the inscription of the kuei vessel made by Ling during the reign of King Wu,111 the founder of Chou as a dynasty, Ling was granted ten p'üng of cowries together with "ten families of ch'ün (subjects) and one hundred li (slaves)." The grant of such a small amount as ten p'üng together with the grant of a large number of "subjects" and "slaves" indicates also the high value of the shells. The inscription on the tsun vessel made by Ch'ien during the reign of King Ch'üng,112 son of King Wu, records that Ch'ien was granted five p'üng of cowries on the occasion when he had a fief bestowed upon him. The inscription on a kuei vessel made by Su during the reign of the same king113 records that Su was exempted from further military service and granted "cowries" for his contribution in the campaign against the Eastern Barbarians. However, the number of p'üng in the grant is not specified. Another minister who was exempted from military service is recorded in the inscription on a ting made by Küng-ying during the reign of King K'ang (trad. 1078—1053 B.C.).114 It is also recorded that in addition to that great favor Küng-ying was granted ten p'üng of cowries. Considering the circumstances under which the grants were made and the fact that the grants, however small in amount, were rewards to highly placed personages who had performed meritorious deeds, five or ten p'üng of cowries must have been something which could match the dignity of both the benefactors and the beneficiaries. Since the cowrie shells were still of considerable value even during early Chou, the generals who captured them in their military campaigns thought the event significant enough to cast bronze vessels in commemoration of the feat.115

In the course of time, the Chinese empire of the Chou period extended its borders further to the south and the southeast. This brought the Chou people closer to the areas where cowries were produced and to the sea coasts whence they were obtained. As a result their import and their supplies of the shells increased. This may account for the grants of cowries in larger amounts of thirty, fifty and one hundred p'üng as recorded in inscriptions of Chou bronzes from the tenth century on. The increase in the amount of reward as the result in the increase in supplies foreshadowed the gradual decrease in the value of the shells. The trend of decrease in value was likely to have continued. Therefore, at the beginning of our era, or in 9 A. D. to be specific, the official value for a shell of the size of, or smaller than, 13.8 mm. in length was worth only three of the "Small Coins" then in circulation. One of the size of, or larger than, no mm. in length was worth no more than 108.116 Four well preserved specimens of the "Small Coins" chosen from the collection of The American Numismatic Society measure 15 mm. in diameter and 1.5 mm. in thickness. Their weights range from 1.34 to 0.98 grams. The comparison of the size and weight of the "Small Coins" with the official value of the cowrie shell will give an idea of the worth of the shells in 9 A. D.

End Notes

94 Wu Ta-ch'eng image K'ü-chai chi-ku-lu (A collection of the ancient inscriptions of the K'ü-chai), 1896, VII, 4.
95 Kuo Mo-jo, Liang Chou chin-wün-tz'ǔ ta-hsi k'ao-shih, 1935, I, 60b.
96 Such as Creel (Birth of China , 92) and Gibson ("Cowries as Money during the Shang and Chou Periods," Journal of the North China Branch of the Royal Asiatic Society, LXXI (1940), 40). Dr. R. S. Britton renders p'üng as "rope" as in the English expression, a "rope" of pearls. (Fifty Shang Inscriptions, 1940, 15).
97 Gibson, Jour. N. C. B. Royal Asiat. Soc., LXXI (1940), Plate VI.
98 L. Wieger, Chinese Characters, their Origin, Etymology, History, Classification and Signification, 2nd print, 1927, 365.
99 Op. cit., 377.
100 H. F. Bowker, "Cowries as Money," The Coin Collector's Journal, X (1943), 92.
101 Shih ching, X, 5a, note.
102 Wang Kuo-wei image (1877—1927), Kuan-t'ang chi-lin, in Wang-chung-ch'iao-hung i-shu, III, 17b—18a.
103 Kuo Mo-jo, Chia-ku wün-tzǔ yen-chiu (Studies of the oracle bone inscriptions), 1931, I, section 10, "Shih P'üng"
104 Creel, Birth of China , 92.
105 Ibid.
106 Chin Yün-o image Hsin-chüng ch'u-t'u ku-ch'i t'u-chih hsü-pien (A supplementary illustrated catalogue of the ancient wares unearthed in Hsin-chüng), 1923, illustration and note on cowries.
107 Yü Hsing-wu, op. cit. III, Part 1, 5a.
108 Pan Ku image (32—92), Han shu, XXIV, Part 2, 12b—13a.
109 Wu K'ai-shüng image Chi-chin wün lu (A collection of bronze inscriptions with comments), prefaced 1932, II, 20a.
110 Jung Küng, Shang Chou i-ch'i t'ung-k'ao (The Bronzes of Shang and Chou), Peiping, 1941, I, 429-431.
111 According to Kuo Mo-jo, Liang Chou chin-wün-tz'ǔ ta-hsi k'ao-shih, I, 3 b.
112 According to Kuo Mo-jo, op. cit., I, 15 b.
113 According to Kuo Mo-jo, op. cit., I, 23a—24a.
114 According to Kuo Mo-jo, op. cit., I, 43b.
115 For a few examples see Kuo Mo-jo, op. cit., I, 25a and 28a.
116 Pan Ku, ibid.



The earliest Chinese coins recorded in ancient literature of which specimens are preserved are the pu image (spade) and the tao image (knife).1 The general belief is that the shape of the pu coin is derived from that of an ancient agricultural tool, which had probably already been in use as a medium of exchange.

Discoveries are known of agricultural implements of a shape similar to that of the pu coin. Kuroda Kanichi reports the discovery of a bronze agricultural tool ( Plate II, 7), hereafter designated as Tool Kuroda, in Chüng-Chou, Honan.2 He calls it a "bronze t'ao," but identifies it with the ch'ien. Obviously he regards ch'ien and t'ao as two names for the same tool. The Tool Kuroda has a socket which extends onto the blade a third of its total length. A ridge runs down the middle of the lower part of the socket to the center of the blade. The upper part or the shoulders of the blade is narrow and rounded while its foot or cutting edge is broad and angular. Kuroda specifically notes that the cutting edge of the blade shows traces of having been used (in weeding or digging).3 Converted into millimeters the measurements he gives for the tool are: length, 181.8; width of blade, 90.9; and thickness, 3.636. It weighs 97 Japanese ryo, or 363.75 grams. Kuroda further observes that it is 21.21 mm. longer than the largest spade coins illustrated in Ku ch'üan hui. 4

In the collection of The American Numismatic Society there is an ancient Chinese farming tool (Plate III), hereafter designated as Tool ANS. It is of bronze and is entirely covered with an aged patina. It also has a socket obviously to accommodate a wooden handle. Like that of Tool Kuroda the socket extends onto the blade and reaches as far as a third of its length. A vertical ridge runs in the center connecting the socket and the blade. Judging from the fact that the ridge starts where the hole of the socket ends, the ridge must have been made for reinforcing the handle. Like the Tool Kuroda, ours also has a hole through the socket, which is triangular in shape and completely pierces it. As suggested by Mr. William L. Clark, Curator of Mediaeval and Modem Coins at The American Numismatic Society, this hole may have been prepared for the insertion of a metal wedge, a wooden pin, a thread, or leather strap to secure the wooden handle in the socket.

On the whole the tool appears rectangular. Its shoulders are square and thick, while its cutting edge is much thinner and somewhat rounded in shape. This must have been the result of long use, which is indicated by the fact that one side of the cutting edge has been worn off much more than the other. The measurements of the tool in millimeters are as follows: length, 123; shoulder width, 84; foot width, 85; thickness of shoulder, 3.5; thickness at foot, 0.5.

Nishimura Shinji 4a has illustrated two farming tools, which, he says, were "excavated in China." One of these is similar in shape to Tool Kuroda, the other ( Plate II, 8) hereafter designated as Tool Nishimura, which is longer, has slightly raised shoulders and a curved cutting edge which is slightly concave. The details of its surface are not clear in the reproduction. There appears to be a socket, though the outlines of that part of the blade are blurred. In all probability its structural features may be the same as those of Tool Kuroda and Tool ANS. Unfortunately, Nishimura does not furnish informative details, nor does he give measurements. In shape it differs markedly from both Tool Kuroda and Tool ANS in that its shoulders are raised and its cutting edge is concave. Although it probably performed the same function as the other tools, it must be a variety from another locality and for that reason, might have a different or local name.

A comparison of the early types of the spade coins with these two types of the ancient agricultural tools reveals a striking resemblance between the money and the implement. The Ku ch'üan hui has a few wood block illustrations from drawings of the early spade coins, which are supposedly reproduced in original size. In volume yüan, pp. 12a and 13a—13b, there are two specimens (we shall designate them as specimens A and B respectively) which are not inscribed, a sign of early origin. As illustrated, both have a socket, which, like that on the tool, extends onto the blade and reaches about a third of the blade's length. The width of the shoulders is slightly narrower than that of the foot. The following are their measurements:

Specimen A Specimen B
Total length 158 mm. 135 mm.
Shoulder width 79 71.5
Foot width 93 84
Socket above shoulder 33 31

Specimen A resembles very much the "prototype spade coin" which was reported in the possession of the Imperial Museum of Japan and is illustrated by Irita Seizo in his article dealing with the origin of the shapes of spade and knife coins.5 (A reproduction of the illustration is found on Plate V, 1). The design of this "prototype spade coin" resembles the features of the ancient tool in every detail. Its measurements, converted into millimeters are: total length, 160.59; shoulder width, 83.931; foot width, 98.475. Its socket (handle) is short, a feature in which it strongly resembles Specimen A. It differs from the latter only in the fact that the edges of the two sides of the blade curve inward and flare out slightly at the bottom.

The American Numismatic Society has an early bronze spade coin (Plate IV), which is in perfect condition save for three small holes which no doubt were incidentally perforated later. Its socket (hollow handle) is short like the one in the Imperial Museum of Japan. The length of the socket on the blade is extraordinarily long, reaching the center of the blade. Its shoulders, which are not round like those of Tool Kuroda, slant at the ends. Its foot (cutting edge) has two pointed tips, which constitute its unique feature. Otherwise, it looks like the other specimens described above. Its measurements are:

Total length 110mm.
Shoulder width 60
Foot width 64
Socket above shoulder 20
Weight 105.1 grams.

The specimens recorded in the Ku ch'üan hut, especially Specimen A, may not have been made as coins. Considering their shape and their large size it is not impossible that they may have been made as tools. However, as the demarcation line between the tool and the coin which gradually developed from it is naturally obscure, any categorical definition as to their specific functions without actual examination of the objects is inadvisable. The specimen in The American Numismatic Society does not present these difficulties. The smallness of its body and the two extra pointed tips at its foot exclude the practicability of its use in the fields. There appears no doubt that it is one of the early, or prototype, spade coins.

As time went on, further changes in design and the size of this type of early Chinese coins occurred. Their size became smaller and smaller, and their body thinner and thinner (see Plates VII—X for examples). Although they retain the socket, it no longer projects onto the blade but ends at the shoulders of the blade. The portion of the socket above the blade is short, whereas that of the coins is comparatively much longer and appears more like a handle. While the blade of the tool is thicker at its shoulders and thinner at its cutting edge (foot), the blade of the coins is of the same thinness throughout. Thus, we see that the spade coins lost all of the functional features of the original spade.

Though the spade coins have functionally deviated from the original tool to a marked degree, their design faithfully preserves the appearance of the tool. All the hollow-handle spade coins bear three vertical lines on both sides. The central line is an imitation of the reinforcing ridge, and the two lines on its sides are reminiscent of the two edges of the extension of the socket into the blade.

This same observation holds true for the flat-shoulder-flat-foot hollow-handle spades (H. H. Spade II) as well as the raised-shoulder-pointed-foot hollow-handle spades (H. H. Spade I), both of which are early coins. The first type seems to be a direct descendant of the farming tools represented by Tool Kuroda and Tool ANS, while the second appears to be that of the tools represented by Tool Nishimura. Thus we may conclude that the pu (spade) coin of ancient China developed out of an agricultural tool from which its shape and design are derived.

Now we must try to explain why the coin was called pu. Since the coin designated as pu developed from an agricultural tool, it is natural to assume that pu must have been the name of this tool. Pu, however, is not found as a name for an agricultural tool in the literature of the Chou period. Apart from being a name for a money the term pu is used regularly to denote a textile. As has been admirably explained by Wu Ch'üng-shih, a famous scholar in Chinese classics, this textile, a material for everday clothing, was made from the fibre of the plant and hemp.6 There is no reason whatsoever for pu as the name of fibre cloth to have become by semantic transfer the name for money.

Although we do not find pu as the name for an agricultural tool in the historical records, we do have the character po image used in this sense. In the Ch'ün-kung ode preserved in the Shih ching (Book of Odes) peasants are instructed to "prepare ch'ien and po" for their work in the field.7 According to Hsü Shün of the first century A. D. "field tools" or "farming tools" were among the meanings for both the characters ch'ien and po in his day.8 His explanations are corroborated by Chüng Hsüan's (127—200) comments on the ode referred to above. Specifically Cheng Hsüan identifies po with lu image a tool for weeding. In the Liang-Ssǔ ode, which is also found in the Shih ching, po is used to signify a weeding tool.9 As the implements which we have discussed above seem to be well suited for weeding we may justifiably identify po with the tool from which the pu coin developed. But why was the coin not called po rather than pu whick is a different character? In this, as in many other instances in Chinese philology, we may find the answer by resorting to one of the basic principles of ancient Chinese linguistics: characters of the same sound can be used interchangeably.

Etymologically speaking, the character po image being signic-phonetic (the radical image for metal is its signic and image or image is its phonetic) is of late origin. Its early or original form is image (pronounced fu today). The form of image as it appears in the bronze inscriptions of the Shang and the Chou periods is reminiscent of the picture of the agricultural tool we have described.10 When later the character po was formed, image became its phonetic, and consequently its basic component part. Because image has imageas its basic (phonetic) element and consequently is pronounced the same as the latter, image could be replaced by image Thus we find that in the inscriptions on the early bronzes image is actually used as a substitute for image 11 For the same reason, in the inscription of the Ch'u-kung Po, image is written image 12 with image substituting for image or image. As we know, image is also the phonetic of the character pu image the name of the spade coin as found in literature. The result is that pu could be used for image the original form of the character po, the name of the agricultural tool. Even at a much later time, in the literature of the Han period, image and image are still used interchangeably. image which is another form for image in the Shi-chi is replaced by image in the Han shu. 13 The name for door decoration is written image and also image 14 Here the "hand" image radical in image and the "metal" image radical in image are auxiliary component parts and are of little or no importance. Since pu image was used for image it could and must have been used for po image As far as we can see, this is probably the way in which pu came to be a substitute for po, which was the name of an agricultural tool and must have been also the name of the coin that developed from such a tool.

It must be noted, however, that in the ode quoted above, ch'ien is mentioned along with po. There is no doubt that ch'ien was also a farming tool, which Chüng Hsüan in his comment on that ode identifies with t'ao image Etymologically speaking, t'ao is also a signicphonetic character, and therefore is late in origin. The character has the same basic structural part image and the same sound as the character image signifying "clam" or "clam shell." In ancient China, the clam shell was also called ch'ün or chün image In remote antiquity the Chinese are said to have "ground clam shells (ch'üin) with which they weeded."15 For this reason both the character lu image meaning a "weeding tool" or "to weed," and the character nung image meaning "farming" or "to farm," have ch'ün as their fundamental component part.16 Thus the farming tool must have been also a tool for weeding. As the name of a weeding tool t'ao is found in many passages in the literature of the Chou period.17 Since t'ao is identified with ch'ien by Chüng Hsüan, the latter is to be understood also as the name of a weeding tool.

What is perplexing is the fact that in the literature of or concerning the Chou period ch'ien, like pu (po), is also found to be a designation of certain money.18 As far as the coins (which are preserved by tens of thousands) show, the money referred to are the spades. Thus we find both pu (po) and ch'ien as appellations for spade coins.

How do we explain this seemingly confusing situation? In order to clarify the confusion in terminology, we must come back once again to the tools and the two different types of hollow-handle spade coins described above. In the description of the farming tools we have shown two different types. One type, represented by Tool Kuroda and Tool ANS, has either slanted or square shoulders and a flat foot (cutting edge). The other, represented by Tool Nishimura, has raised shoulders and a concave foot. The existence of the two different shapes of tools is paralleled by the existence of two different shaped hollow-handle spade coins which were developed from these tools. One type of the hollow-handle spade coins (designated as H. H. Spade I in the following text) is marked by raised and pointed shoulders and pointed feet (see Plate VI), and the other (designated as H. H. Spade II) is characterized by flat or square shoulders and flat or mildly curved feet (see Plates VII—IX). There is no indication or possibility that one type was a variety of the other. These two differently shaped early spade coins must have been local varieties developed from two differently shaped tools. This point is confirmed by the discovery of the ancient tools described above, whose characteristic features can still be seen in the designs of the coins. One type of tool must have been called ch'ien from which the spade coin designated as ch'ien developed, and the other must have been called po (pu) from which the spade coin designated as pu developed. Only by such an understanding of the monetary terms, can we comprehend the statement of Ssŭ-ma Ch'ien that ancient Chinese coins, as far as the spades are concerned, "were either the ch'ien or the pu" 19

To determine which type was the po (pu) and which the ch'ien is difficult. There is, however, indirect evidence which may be of some aid in solving the problem: the names and forms of the musical bells of the Chou period. The bells which are preserved are of two types which when viewed from the side appear in the shapes image and image. They are designated as chung image and po image respectively.20 The muscial bell po may have been so called because of its resemblance to the farming tool po.

The distinctive features of the two types of spades attracted the attention of Ch'ün Chieh-ch'i, a noted numismatist of the last century who remarked that the spades with a flat foot are called po and that those with a pointed (or, in his words, "not flat,,) foot are called ch'ün.21 He also refers to the shape of the po bell as evidence. If Ch'ün Chieh-ch'i's observation is correct, then Tool Nishimura, from which the raised-shoulder-pointed-foot spades appear to have developed, must have been called ch'ien, and Tool Kuroda and Tool ANS, from which the flat-shoulder-square-foot spades appear to have been developed, must have been called po. It is significant that when the term ch'ien had become an exclusive name for money, the agricultural tool which probably was originally called ch'ien was designated solely as t'ao. At the same time the term pu, which was originally a borrowed character for po, a tool, became a special term for the spade coin. All indications point to the fact that because of the growing importance of the tool-money and because of the increasing deviation in their design from the original tools, there seemed to have been a tacit practice among the ancient Chinese to give these money types exclusive appellations. Thus ch'ien and pu became the conventional terms for the tool-money, and t'ao and po for the tools.22 In one case the tool was given a completely different name, and in the other the money appropriated the borrowed term, pu. In a way this terminological evolution for the spade coin illustrates the early history of Chinese coinage.23

One more point can be noted in connection with the terms ch'ien and po. At the end of the Chou period, ch'ien seems to have been used in a much broader sense than pu. While pu designates only the spade coin, ch'ien appears to denote money of all types, including the spade. On account of this, when the round coins appeared in the third century B. C. they were also called ch'ien, but never pu. Today, ch'ien is still being used to mean money in general.24

End Notes

4a Nishimura Shinji, "Ancient Chinese Coinage and its Origin," Canton, Vol. I, (1939) 34.
1 Yüan-chin or "metal plate," the money of the state of Ch'u during the Chou dynasty, is not recorded in historical literature. For discussion of Yüan-chin, see pp. 180—186.
2 Kuroda Kanichi image "Shūdai no kinzoku kahei ni tsuite," (Coole JM-10,0.), Kōkogaku zasshi image XVI (1926), 138.
3 Op. cit. 139.
4 Ibid.
5 Irita Seizō image "Tōfu no keishiki to sono kigen" (Coole, JM-10,k.), Kōko-gaku zasshi, XV (1925), 402—412.
6 Wu Ch'üng-shih image Pu po ming wu, 1930, la—3a.
7 Shih ching (Mao Shih), XIX, 6b.
8 Po means also the ornamental carvings on the horizontal stick at the top of the bell.. Ch'ien means also money. Tuan-shih shuo-wün-chieh-tzǔ chu, 1908, XIV, 3a and 4b.
9 Shih ching, XIX, 15a.
10 For the forms of the character in bronze inscriptions, see Jung Küng, Chin wün pien III, 37b.
11 So states Jung Küng, loc. cit.
12 Kuo Mo-jo, Liang Chou chin-wün-tz'ǔ ta-hsi t'u-lu, 1934, 177.
13 Shih-chi, CXVII, 16a and Han shu, LVII, Part 1, 9b.
14 Hsü Shün states that image is the image attached to the door. (Tuan-shih shuo-wün chieh-tzǔ chu, XIV, Part 1, 7a). The T'ung-su wün image states that the decoration of the door is called image (T'ai-p'ing Yü-lan, 1818, Pai ed., CXXCVIII, 4b). The latter term is also found in Han shu, 1644, XI, 5a.
15 Huai-nan-tz ǔ, XIII, 1b.
16 For a fuller discussion of this problem see Kuo Mo-jo. Chia-ku wün-tzǔ yen-chiu, 1931, II, 25a—26a.
17 Practically all statements regarding the t'ao in Chou literature have been collected by Kuei Fu image in his comments on the character in Shuo-wün chieh-tzŭ i chüng, 1870, XLV, 14a—14b.
18 Chou shu image quoted in the T'ai-p'ing yü-lan, CCMXXXV, 1b; Kuo-yü, III, 10b; Han-fei-tzŭ, 1875, XI, 9a; Lü-shih ch'un-ch'iu, 1875, XV, 3b; Shih-chi, III, 10b.
19 Shih-chi, XXX, 21a.
20 Both T'ang Lan and Kuo Mo-jo hold the opposite opinion. What we call po they call chung, and what we call chung they call po (see T'ang Lan, "Ku yüeh-ch'i Hsiao-chi," Yenching Hsüeh-pao Nimageo. 14 (1933), 82—83; and Kuo Mo-jo, Liang-Chou chin-wün-tz' ŭ ta-hsi t'u-lu, "T'u shuo," 4b—5a). Actually the appellations of the music bells has been a problem of uncertainty for many centuries. This uncertainty seems to have grown out of the confusion with which the terms were used when some of the late bells were cast. At this time the people appeared to have forgotten the original distinction between the two types of bells as implied in their two designations, chung and po, and some named the bells indiscriminately. This is probably the reason which led Jung Küng to the conclusion that there are no differences between the po and the chung (Shang Chou i-ch'i t'ung k'ao, I, 495).
21 Quoted by Pao K'ang, Kuan-ku-kü ts'ung-kao san-pien (Coole 299), 1876, Part 1,6a.
22 Po as an agricultural tool still appears several times in the K'ao-kung chi, the last section of the present text of the Chou li, which is most likely a work of the Chan-kuo period.
23 There are two other explanations of the origin of the monetary term pu, which, however unconvincing, must be noted. One theory, advanced by two Japanese scholars, is that the term pu evolved from the name of the ancient Chinese ax, which is chin image according to Kuroda Kanichi(Kuroda, Kōkogaku zasshi, XVI (1926), 141) and fu image according to Tanaka Keibun quoted by Nishimura Shinji (Canton, Vol. I, 30). The reason Tanaka identifies pu with fu is that fu has a "common origin" with po image, from which the pu coin developed. Nishimura's quotation of Tanaka's explanation may not have been sufficiently adequate. As it is, we cannot understand why Tanaka should approach the problem indirectly through fu to po and not directly as we have. What caused Kuroda to trace the origin of the term pu to chin is that the character chin appears in the legend on Old Spade coins (those coined after the hollow-handled spades). The difficulty with Kuroda's explanation is that philologically speaking pu has nothing to do with chin (ax), and he does not seem to have found it necessary to offer an explanation.
The other theory advanced by Chüng Chia-hsiang, a Chinese numismatist, in his article on the origin and evolution of monetary terms (Ch'üan-pi, No. 22, 6—9), is that the spade coin was named pu (which term he understands as cloth) because the royal treasury of Chou had collected cloth as its revenue. He concedes, however, that his interpretation has no bearing on the original name of the coin.


The dating of the beginning of Chinese coinage is one of the most controversial problems in Chinese numismatics. Ssŭ-ma Ch'ien (145—86 (?) B. C.) places its appearance as early as the legendary Yü and the semi-legendary Hsia dynasties.25 Some Sung scholars attributed the spade coinage to even more remote pre-historical figures. Though modern numismatists and historians are more realistic, their opinions diverge considerably. Our own immediate concern is with the dating of the spades. Of the various theories advanced concerning them, the following may be regarded as representative.

1. Hsia and Shang dynasties (ending 1122 B. C., traditional date). Among the modern numismatists Lacouperie appears to be of this opinion.26 Though he does not categorically commit himself to such an early date for Chinese coinage, one gets this impression from his introductory remarks on spade coinage. As a specimen of the spade coinage of this period he gives a hollow-handle spade with raised or pointed shoulders and pointed foot.

2. Western Chou (1122—771 B. C.).

Numismatists and historians who subscribe to this opinion are many. Chüng Chia-hsiang may be taken as their representative.27 He arrives at this date from three considerations. First, the Western Chou had already had a "fully developed governmental organization." "Because of the necessity to collect pecuniary fines and taxes, copper money was made." Secondly, the epigraphical style of the coin legends on the spades is that of the so-called "great seal character," which is the style of script of the Western Chou period. Thirdly, the money which was used during the Western Chou period was the ch'üan which were the spade and the knife coins.

3. End of the Ch'un-ch'iu period (770—481 B. C.).

Ojima Sukema may be regarded as the exponent of this theory.28

His basic arguments are: (1) that the historical records of or concerning the Western Chou and the Ch'un-ch'iu period show that a natural economy prevailed at that time; (2) that literary sources show that a money economy did not develop until the Chan-kuo period (403—221 B. C.). The literary sources he used are Tso chuan, Kuo yü and Chan-kuo ts'ü. Though Kuo yü contains a story that King Ching of Chou made "big coins" in 524 B. C., he doubts its authenticity.

4. The Chan-kuo period (403—221 B. C.).

The scholar who holds this opinion is Li Chien-nung. In his article dealing with the development of the monetary system before Ch'in (221—207 B. C.)29 he observes that as late as the Ch'un-ch'iu period the Chinese still practiced a barter economy or "exchange in kind." Therefore he states that spades and knives became circulating currency after the commencement of the Chan-kuo period. The historical materials which he used were those used by Ojima. He also rejects the story of the casting of "big coins" by King Ching as unhistorical.

The faulty logic of some of Chüng Chia-hsiang's arguments and his error in interpreting the meaning of ch'üan are obvious.30 It may be mentioned in passing, however, that he is the only one of the above four who made use of any evidence from the coins themselves. Whereas Lacouperie used one coin specimen to illustrate his point, that specimen, being late in date, had no actual bearing on his theory. He and the others based their conclusions entirely on literary evidence.

Basically, the ancient historical literature of China is political in nature and is generally terse inform; it contains little regarding contemporary economic life. It has not been completely preserved; what we have of it today is in fact only a small part. Therefore, any insistence on the non-existence of things unmentioned in such a limited amount of annalistic records is bound to be untenable, and, in many cases, untrue. For example, the Ch'u state of the Chou period was never reported as having money of its own. Today, however, we have Yüan-chin money which has been unearthed in central China.

In their use of the literary sources, the scholars have, moreover, failed to observe the rudimentary principle of first examining the nature of the data. Li Chien-nung states that he selected about eighty passages from the Tso chuan which refer to "forms of property" in none of which is there mention of either spade or knife coins.31 According to Li, twenty of these passages concern bribery, twenty-nine awards and gifts, nine offers in expectation of favors asked for, nine seizure of one noble's property by another, and seven wealth in general. Significantly, however, Li Chien-nung fails to note either the status of the men who are involved in the events or the nature of the events related in his passages. We may say briefly that the objects which he calls "forms of property" are limited to those which could be exchanged among the nobles as presents and the possession of which alone was regarded at the time as respected symbols of wealth. Thus the usual objects of the grants of the king to the feudal lords and the presents exchanged among the latter are jade, silk and silk embroidery, carriages and horses, slaves, gold. When the wealth of a noble is mentioned, it is expressed in terms of the number of horses and carriages he used in his equippage. How much money was in his treasury and how much grain in his granary were matters of secondary importance. The objects which one noble lord tried to seize from his rivals in war were territory, crops, weapons, war horses, and sacrificial bronze vessels — the symbol of the authority of a state. Jade has been mentioned as trophy, but it is always an important or famous piece, the possession of which has been too often a cause of conflict. Even as late as the Han time the imperial grants which were given actually in copper cash were usually expressed in terms of gold. To give money as a present is still frowned upon by the Chinese. Any suggestion that the objects offered as gifts in an aristocratic society are the only "forms of property" and thereupon that money does not exist is largely due to an inadequate understanding of the social and psychological aspects of the property valued by the nobility.

The Tso chuan, moreover, which is relied upon by both Ojima and Li Chien-nung as their authority for the denial of the existence of money during the Ch'un-ch'iu period is not completely devoid of the mention of the spade coin. A passage in the Tso chuan 32 relates that in 517 B. C. the prince of Lu, Duke Chao, was forced to flee to Ch'i by three noble families who competed with him for power. In the next year, 516 B. C., Duke Ching of Ch'i tried to return Duke Chao to his state and instructed his minister Tzŭ-yu, who was to carry out the plan, not to accept bribery from the nobles of Lu lest the plan be sabotaged. Two officials from one of the three families which expelled Duke Chao promised Kao I, an assistant to Tzŭ-yu, generous political and property compensations if he could bribe and dissuade his superior from carrying out the order. Kao I showed the silk embroidery which he received from the emissaries of the noble family of Lu to Tzŭ-yu, and he desired to have it. Then Kao I said to Tzŭ-yu, "To buy this the Lu people pay one pu (spade coin) for one hundred bolts." indicating that the noble families of Lu could procure with ease (at a very low price) large quantities of the silk embroidery to be offered to Tzŭ-yu if he should help them realize their wishes. Kao I's statement in quotation marks is a verbal translation of the Chinese text, and the only acceptable one possible from the grammatical construction of the sentences and the meaning implied therein. While Hui Tung (1697—1758) and Hung Liang-chi (1746—1809), both of whom are famous scholars in the field of Chinese classics and history and have studied the Tso chuan in particular, agree with our interpretation,33 Shün Ch'in-han (1775—1831) finds it objectionable on the ground that the price for the silk embroidery is too low.34 He offers an interpretation in which the character pu means tsu (accumulation, pile). In the end this interpretation agrees with the one advanced by Tu Yu (222—284), who understands pu as meaning ch'ün (to display, to exhibit).35 In his translation James Legge renders the passage thus, "the people of Loo (Lu) had bought such silk, made up in 1000 pieces."36 This constitutes a third interpretation. But his translation does not agree with the Chinese text. Tu Yu and Shün Ch'in-han's explanation is unnatural and incongruous with the idea which the author tries to convey. The objection of Shün Ch'in-han to our interpretation would have some validity were it not for the fact that the question of price is not important. Understandably, Kao I's statement must have been grossly exaggerated in his eagerness to convince his superior that bribes could be provided easily and in large quantities."37In the text, "to buy (mai)" is the action, and "the spade coin (pu)" is the means with which the action of buying is consumated. The idea is clearly expressed. It is difficult to understand how Li Chien-nung, and for that matter, Ojima too, missed the passage.

If they have missed the mention of pu money in the Tso chuan, they should not have overlooked that in the Li chi, a Chinese classic. There pu money is mentioned in two passages as funeral gifts. These passages are found in the T'an-kung chapter which is an authentic historical record for the Ch'un-ch'iu period. In both cases the authoritative Han commentator Chüng Hsüan (127—200) specifically notes that pu was money.38 Judging from their failure to use these important historical data, we can but draw the conclusion that Ojima and Li Chien-nung have not exhausted the very source of information on which they have relied in their studies.

As a matter of fact the spade (pu) as money appeared in a much earlier record. The record is the ode of Mang in the Shih ching. In this ode a girl sings of her suitor

A simple-looking lad you were, Carrying pu (spades) to exchange for silk. But you came not so to purchase silk; You came to make proposals to me.39

In date this ode cannot be later than the sixth century B. C. The character pu in this ode is given by both Chüng Chung and Chüng Hsüan of the Han dynasty as meaning pi (money).40 But some modern scholars, for reasons not specified, have disregarded these early commentaries and interpreted it as meaning "cloth." In their translations of the ode James Legge, Author Waley, and Bernard Karlgren express the same unwarranted opinion.41 Ojima and Li Chien-nung subscribe to this explanation, and use it to prove their thesis that at the time the ode was composed there existed an economy of "exchange in kind."

In the discussion above we have demonstrated that the term pu was used as a common noun for cloth or textile; it denotes, however, only the cloth made of grass fibre such as the plant and hemp. (See p. 94). Such cloth is inferior in quality and coarse in workmanship. If it is used as a denominative meaning a kind of textile as expressed in the term tsui pu (felt), it also implies an ordinary material. If the love-seeking boy who pretends to be a merchant had brought cloth with him to exchange for silk, he could not have "carried it in his arms in front of his chest" as the verb pao implies. It would be not only cumbersome but also inconvenient to carry the large amount necessary to purchase a much finer and more treasured textile, silk. But if we follow the commentary of the Han scholars who were closer to the old tradition than modern scholars and understand the term pu as meaning spade coins, then there is no difficulty in understanding the verb pao.

According to Lo Chün-yü, an astute collector of archaeological objects and ancient coins, some of the unearthed hollow-handle spade coins which he saw were fastened together at their handles with a cord.42 Kuo Pao-chün reports to the author that the hollow-handle spades which he discovered in the Chou tombs in Chi County, Honan, were also fastened together, though the number in a bundle varies. In actual use, the spades might have been kept together in the same manner. In a story recorded in the Han-fei-tzŭ, a woman in the state of Wei, where the ode of Mang originated, is said to have plead with Heaven that she be granted "one hundred bundles (shu) of spades (pu)."43 This story implies not only that spade coins were bound together in actual use but also that there were a definite number in each bundle.

The modem scholars' unwarranted disregard for early commentaries and their misunderstanding of pu in these particular cases as meaning "cloth" is further illustrated by their explanation of a passage in the Mo-tzŭ. In this passage Mo Ti (480—390 (?) B. C.), a philosopher, is quoted as saying, "The gentlemen (shih) of today in making available their personal services take even less care than the merchants in putting to employment i pu (one spade)".44 The meaning of the phrase i pu is so clear that no other interpretation is possible. However, unfamiliar with the ancient Chinese coinage and doggedly believing that the people of late Chou still used cloth as a medium of exchange, some modern scholars read pu in this passage also as meaning "cloth." Equivocation may be tolerated in a direct quotation in Chinese of this passage, but in translating it into a foreign language such as English, one must be specific. Facing this difficult situation Yi-pao Mei interpolates the original text rather than seek for a correct interpretation; in his translation of the passage, he has inserted the word "bale" between the two characters of the phrase i pu, and rendered it as "one bale of cloth."45

Taking the ode of Mang as our authority, we may say that, as far as literary information on the spade goes, the coin must have become an ordinary medium of exchange before the sixth century B. C. or earlier. But this date is not to be taken as the date of the commencement of the spade coinage which in all probability must have been much earlier.

In his statement on the early history of Chinese coinage Ssŭ-ma Ch'ien (145—86? B. C.), China's earliest great historian, says, "With the opening of exchange between farmers, artisans and merchants, there came into use money of tortoise shells, cowrie shells, gold, the ch'ien spade, the knives, and the pu spade. This has been so from remote antiquity."46 A statement such as this is too general to be of any use for numismatists. In his historical account of the Chinese monetary system Pan Ku excludes the Hsia and the Shang periods for lack of information, and starts with the Chou dynasty 47He fails also to give any specific date. Since the Chou period covers some eight hundred years, it is too long a period to be spoken of in general terms. Stories such as those concerning Yü of the Hsia dynasties and T'ang of the Shang dynasty, who are said to have coined money in the second millenium B. C., lack historical authenticity and consequently must be discarded.

In the absence of literary information we must resort to archaeology which, unfortunately, for the reasons we have stated in the introduction, offers little help. The only spade coins which were scientifically recovered are those found by Kuo Pao-chün (See Plate VII, 1). He dated the tombs in which these spades were discovered as of about 270 B.C. However, as Mr. Kuo points out, the spades found are apparently mortuary money especially made for the purpose of burial. They are extraordinarily thin, small and without legend (monetary spades of the same small size are of late origin and always bear a legend). As mortuary money they were not necessarily modelled in shape and design on contemporary money. More likely they represented a money of days long past like the paper horse-shoe silver burnt by modern Chinese on sacrificial occasions which imitates silver ingots long out of circulation.

Of the many thousand bronze inscriptions of the Chou period, we find three have the character pu. In one, the Shou-kung Tsun inscription, it appears as the component part of the term tsui pu, a felt made of animal hair. In the other two it appears as an independent term and may mean spade coins. One of these is found in duplicate on the body and cover of a yu vessel made by Yüan.48The other is found on a tsun vessel also made by Yüan.49 With its first line lost the rest of the inscription varies only slightly in wording from the inscription on the yu vessel. There is no doubt that both the yu and the tsun vessels and the inscriptions thereon were made by the same Yüan as memorials for the same event. The inscription on the yu vessel reads:

In the nineteenth year, the King was in Han. Wang Chiang (the queen) ordered Scribe Yüan to pacify the ruler of the I (barbarians). The ruler of the I presented to Yüan cowries and pu as a gift. To show his gratitude to Wang Chiang's favor he (Yüan) made this precious vessel in honor of his father Kuei.50

Yü Hsing-wu, so far as we know, is the only epigraphical scholar who has commented on the character pu, and he lists two meanings for it, p'i po or "bolt of silk" and pi or "money."51 In Chou literature pu is susceptible to two explanations, cloth or spade coin, depending on the context. As a term for cloth it denotes the coarse cloth made of and hemp only. As late as the Han time it still denotes a cloth made of i52 which is hemp fibre.53 Furthermore, in Chou bronze inscriptions the silk fabrics granted by the king or by a noble to his inferior are either called ssŭ (silk) or po (silk fabrics)54 and nowhere are they called pu. We may, therefore, reject the interpretation of pu as meaning silk fabric.

Can we then regard the character pu in this inscription as signifying fibre cloth? This seems also unlikely. The inscription states clearly that the donor of the cowries and the pu is a ruler of a state even though the state is of a non-Chou origin55 The recipient of the presents is a Scribe (tso ts'ü), who, as one in charge of composing and transmitting royal decrees, was an important official of the court of Chou. It seems rather improbable that the presents offered by a state's ruler to a high minister of the court of Chou would consist of ordinary fibre cloth. If it had been a special kind of cloth, worthy of conveying special respect and deserving special consideration, it would have been recorded by its special name. There seems to be no satisfactory interpretation other than the explanation that pu in this inscription means spade money. A present consisting of spades and cowrie shells, which were also money at the beginning of the Chou period, makes sense. Furthermore, pu as a monetary designation for spades is borne out not only by the literary records but also by the legend on the Fün spades. The two known specimens of these are illustrated in Okutaira, IV, 52 b, and in Fang Jo's Yüeh-yü ku-huo tsa-yung (reproduced on Plate XV, 2). Their legend reads Fün pu or "Pu of Fün." Pu here can refer to nothing else than the spade.

The date for this inscription or the vessel on which it is inscribed has been suggested by Wu Ch'i-ch'ang as the nineteenth year of King Chao,56 the fifth king of Chou if we should regard King Wün as the titular founder of the dynasty. As Kuo Mo-jo has pointed out, the facts implied in the inscription, especially the name Chiang of the queen, do not permit such a date. Therefore he suggests the "nineteenth year" recorded in the inscription is that of the Chou as a dynasty (see below).

As students of ancient Chinese history all know, the chronology of early Chou, that is before 841 B. C., is very uncertain, and in fact has been a controversial topic ever since the Han time. The problem is too complicated even to touch upon here. Prof. Tung Tso-pin lists fifteen different authorities on the subject, who hold eight different opinions.57 He himself is of the opinion that the Chou period officially begins with the following year after King Wu succeeded his father King Wün as the ruler of Chou in 1122 B. C. and actually begins with the year after King Wu conquered the Shang dynasty, which took place in 1111 B. C.58 In the main he follows the traditional date. Dr. R. S. Britton told the author that as far as the records of the moon eclipses on the oracle bone inscriptions of the Wu-ting period go, Tung's date is untenable. He regards as more probable that the beginning of Chou was around 1027 B. C., which date has been suggested by Lei Hai-tsung and B. Karlgren who base their opinion on the chronological record in the original Bamboo Annals (Chu-shu chi-nieri) and other relevant historical data.59 For our present purpose we may follow the traditional date of 1122 B. C. as the beginning of the Chou period with the understanding that the correct date may be a century later.

In his study of the bronze inscription quoted above Kuo Mo-jo follows Wang Kuo-wei's chronology.60 Wang Kuo-wei has proved that the recording of the chronology of the early years of Chou is not according to the reigns of the kings as is the case later. The recording of the year goes on with one serial number starting with King Wün's reign until the seventh year of the reign of King Ch'üng, the third king of Chou. In his opinion King Wün ruled seven years and was succeeded by King Wu who conquered Shang in the eleventh year of Chou or the fourth year of his own reign. King Wu was succeeded by King Ch'üng, whose reign officially begins in the fourteenth year. So calculated, the "nineteenth year" mentioned in the inscription falls in the sixth year of King Ch'üng's reign, which corresponds to the conventional date 1110 B. C.

The presentation of a gift of cowries and spades by the ruler of the I people to Yüan, the royal scribe, appears to have taken place either in Han or in its neighborhood. Kuo Mo-jo identifies Han with the original territory of Han Cho or Cho of Han, and located it within the boundaries of the present county of Wei in eastern Shantung.61 Both his identification and the location of the place are plausible, when we take into consideration the fact that by "I" the Chou people referred to all people in the eastern part of ancient China who were either Shang or their subjects. After the third year of King Ch'üng, the Chou court started a large scale military campaign against such eastern states as Yen and Po-ku, which were either of Shang people or their vassals. These states were located in present Shantung. The campaign, which lasted three years, resulted in the subjugation of those states. It is possible that at the end of the victorious campaign King Ch'üng was brought to the newly conquered territories to establish the Chou authority. This may serve to explain the statement "the king was in Han."

If all these interpretations are correct, the inscription has considerable importance in the dating of the spade coinage. For, if during the reign of the third king of Chou, which by the conventional chronology is in the last years of the 12 th century, or, at latest, in the last years of the 11th century, spade money had been offered as a gift, the beginning of the coinage must have been somewhat earlier. This would lend credit to the hypothesis that spade coinage was a Shang invention.

In this connection we may introduce some records to the effect that the last Shang king had stored spade money at the Lu-t'ai, his treasury. The Chou shu (Book of Chou) states that when King Wu conquered Shang he took out and distributed among the peopel "the ch'ien (spade coins) of the Lu-t'ai" and "the grain of the Chü-ch'iao (a granary)" of the Shang king.62 Ssŭ-ma Ch'ien relates that Ti-Hsin, the last king of Shang "overtaxed [his people] to increase [the storage of] the ch'ien at the Lu-t'ai."63 The distribution of ch'ien from the royal treasury of the Shang by King Wu of Chou is also recorded in several other old literary sources.64 As has been explained before, ch'ien is but another type of spade money or another name for it. Historians have been duly cautious in not giving full credit to these statements, but, in view of the record of spade coins in the early Chou bronze inscription, they may well contain historical truth.

At this juncture we may recall that in the section dealing with the early history of trade in ancient China, we have noted that the Shang period may have witnessed considerable commercial activities. Under these circumstances, the coinage of a metallic money is not at all beyond possibility. However, we must repeat that our suggestion of the possible early Chou or late Shang origin of the spade coinage depends largely upon our interpretation of the character pu in the inscriptions on Yüan's vessels. We must safeguard ourselves against any definite conclusion based on a single piece of evidence, which may some day prove unacceptable.

End Notes

24 Related to the problems of early monetary terms is another controversial subject which deserves a few lines of mention, that is, which term is earlier, ch'ien imageor ch'üan image.The latter term means a fountain or a spring. As early as 9 A. D. the emperor of the Hsin dynasty, Wang Mang, had borrowed it and used it in the sense of money, and had had it cast in the legends on both his "Small Coin" and the "Big Coin," but the term had lost its literal usage with the end of his reign. Elder Chinese numismatists who seem to relish anything uncustomary (the term ch'ien has been used as the general term for money ever since the Chou time) prefer ch'üan to ch'ien as the name for old coins. Therefore, for the titles of coin catalogues we have Ch'üan chih, Ku ch'üan hui etc. But the novelty devised by Wang Mang has produced a belief in the mind of many that as a monetary term ch'üan might be older than ch'ien. Many Chinese numismatists, led by Li Tso-hsien, cherish this idea. A few others do not, and hence the endless debate, which lasts until today. Readers interested in this dispute can find information on it in the various works on Chinese coins, especially in the Ch'üan-pi (Chinese Numismatics, Nos. 4, 21, and 22). It is unnecessary to repeat the argument here. In our opinion the term ch'ien is earlier than ch'uan. The latter was first used by Wang Mang. ch'ien was the name of an old farming tool from which the spade coin developed. It is only natural that the spades, at least one group of them or those in one particular region, were so called. It is true that the character ch'üan meaning money appears in the text of the Chou li, and this has been taken as a proof for its early origin. But, as Ting Fu-pao has pointed out, the character ch'üan in the Chou li, is a later revision made during Wang Mang's time (Chinese Numismatics, No. 4, p. 27), as in an older edition of the work which was seen by Chüng Chung (first century A. D.), the earliest known commentator of the Chou li, ch'ien is found in lieu of ch'üan. The reason Wang Mang adopted ch'üuan for ch'ien was his dislike for the character chin (metal) which is the component part of ch'ien and liu image, the name of the ruling house which he usurped. Technically, ch'üan and ch'ien sound similar in their pronunciation (Liang Ch'i-ch'ao image (1873—1929) called attention to this many years ago in his article "Chung-kuo ku-tai pi-ts'ai k'ao," Yin-ping-shih wün-chi 1916, XX, 70), and this made it possible to substitute easily the former for the latter.
25 Shih-chi, Pa-na ed., XXX, 20b.
26 See his Catalogue of Chinese Coins, London, 1892, 1.
27 Chüng Chia-hsiang image "Shang-ku huo-pi t'ui-chiu" (An investigation into the monetary systems of ancient China), Ch'üan-pi (Chinese Numismatics Bi-monthly), No. 3, pp. 26—28.
28 Ojima Sukema image "Keizai-jō yori mitaru Shōsho no shokukei," Shinagaku image I (1921), 420—436, and "Shunshuū jidai to kahei keizai," op. cit., 539—545, 626—641.
29 Li Chien-nung image, "HsienCh'in huo-pi chih-tu yen-chin k'ao" (A study of the development of the monetary systems of the pre-Ch'in period), Shü-hui k'o-hsüeh chi-k'an image (Social Science Quarterly, National Wu-han University), III (1933), 481 —509.
30 Li Chien-nung, 501.
31 Li Chien-nung, 499 —500.
32 Tso chuan, LII, 1b— 2a. Legge, Chinese Classics, 1872,V, Part II, 712, 716. Instead of "Tzŭ-yü" image and "Kao I" image the Shih-chi has "Tzŭ-chiang" imageand "Kao Hü image respectively (XXXIII, 19a). The differences are probably due to copyist's mistakes, but which versions are correct is not known.
33 Hui Tung image (1697—1758), Ch'un-ch'iu-tso-chuan pu-chu, Huang-ch'ing ching-chiai (Hsüeh-hai-t'ang) ed., CCCLVIII, 2b; Hung Liang-chi image(1746—1809) Ch'un-ch'iu-tso-chuan ku, Ssŭ-pu pei-yao ed., XVIII, 9a.
34 Shün Ch'in-han image (1775 —1831), Ch'un-ch'iu-tso-shih-chuan pu-chu, Ts'ung-shu chi-ch'üng ed., X, 197.
35 Tso chuan, LII, 2a.
36 Legge, ibid.
37 The numerals "hundred (po)" and "one (i)" in the text may be incorrect substitutes for "five (wu)" and "ten (shih)" as in the old script the latter two could be mistaken, for the former two. In either case the price would be considerably reduced.
38 Li chi, 1871, VIII, 1a and 16b.
39 Shih ching (Mao shih), III, 11b—12a.
40 Ibid.
41 Legge, Chinese Classics, 1871, IV, 97; Waley, Book of Songs, London, 1937, 96; Karlgen, "Book of Odes," BMFEA, No. 16 (1944), 190.
42 Lo Chün-yü, Yung-lu jih-cha (Coole, 392), 17a—17b.
43 Han-fei-tzŭ, 1875, X, 4b.
44 Mo-tzŭ, Ssŭ-pu pei-yao ed., XII, 3a.
45 Yi-pao Mei, The Ethical and Political Works of Motze, London, 1929, 225.
46 Shih-cki, XXX, 2, 1a.
47 Han shu, XXIV, Part 2, 1a.
48 Fang Chün-I image Chui-i-chai i-ch'i k'uan-shih k'ao-shih (Studies of the inscriptions on the bronze vessels of the Chui-i-chai), Photostated, XII, 9b.
49 Huang Chün image Tsun-ku-chai so-chien chi-chin T'u-lu (Tsun-ku-chai illustrated catalogue of the bronzes), 1936,1, 36a.
50 We follow the transcription of Kuo Mo-jo, Liang Chou chin-wün-tz' ŭ ta-hsi k'ao-shih, 1,14a.
51 Yü Hsing-wu image, Shuang-chien-ch'ih chi-chin wün-hsüan (Shuang-chien-ch'ih selection of the bronze inscriptions), 1933, III, Part 3, 10a.
52 Shuo-wün chieh-tzŭ, VII, Part 2.
53 Ibid.
54 For examples see the inscriptions of the T'ung Kuei and the Shou-kung Tsun in the Liang Chou chin-wün-tz' ŭ ta-hsi k'ao-shih, II, 87b and 92b.
55 I was the general designation of the Chou people for the Shang people and the peoples who lived in the eastern part of ancient China. See Fu Ssŭ-nien image "I Hsia tung hsi shuo," Ch'ing-chu Ts'ai Yüan-p'ei hsien-shüng liu-shih-wu-sui lun-wün-chi (Studies presented to Ts'ai Yüan-p'ei on his sixty-fifth birthday), Peiping, 1935, 1093–1134.
56 Wu Ch'i-ch'ang image, Chin-wün li-shuo shu-chüng (A study of the chronology of the bronze inscriptions), 1936, II, 30a.
57 Tung Tso-pin image, Yin li p'u (Calendar of Yin), 1945, Part I, IV, 12a.
58 Tung Tso-pin, op. cit., 22b.
59 Lei Hai-tsung image "Yin Chou nien-tai k'ao," Wün-che chi-kan image (Wu-han University), II (1931—2), 1 —14. B. Karlgen, "Some Weapons and Tools of the Yin Dynasty," BMFEA, No. 17 (1945), 120.
60 Wang Kuo-wei, "Chou k'ai-kuo nien-piao" (The chronology at the beginning of the Chou dynasty), Kuan-t'ang pieh-chi pu-i, in Wang-chung-ch'iao-kung i-shu, 4a—8b.
61 Kuo Mo-jo, op. cit., 14a and 16a. The possibility that spade coins were used at a very early date in the general area around modem Wei county is evidenced by the existence of a Prototype Spade coin and several Hollow-handle Spades cast by the city or mint of I image ( Plate X, 1). This I is identical with the I image city which cast the round coins of four denominations at the end of the Chou period. The lower component part of the latter I is a later addition, the presence of which does not bring about any change in the meaning of the character. The I which cast the round coins has been located in present I-tu county whose county seat is about thirty miles west of that of Wei county in eastern Shantung. Some time after Chou conquered or subjugated this general area, it abandoned the spade coinage and adopted the knife coinage the origin of which took place probably in a state further to the east (see p. 156).
62 Quoted in the T'ai-p'ing yü-lan, photostat of the Sung edition in the Ssŭ-pu ts'ung-k'an ed. CCMXXXV, 1b.
63 Shih-chi, III, 10b.


Attempts have been made in the past by a few to trace the evolution of the various types of spade coins.65 Their observations, however, have been too superficial to be of much value. Since scientific reports on discoveries of spade coins are lacking, a reconstruction of the evolution of their designs remains largely conjectural. We proceed with the conviction that the later the design of the coin, the less its resemblance to the original tool; and that the later the coin, the smaller its size and the lighter its weight. The gradual reduction in size and weight of coinage is a phenomenon common to the historical coinages of many peoples, and Chinese coinages could hardly have deviated from this law.

A number of spade coins bear no legend, but many more do. Judging from their design and weight, we may say in general that those without legends are older, and that those with legends are later. Among the spades with legends, those having a numeral or a character from the "heavenly stems" (numbering ten in all) or the "earthly branches" (numbering twelve in all) as their mark, or "serial mark," are earlier than those which have the name of a mint and, as frequently, the name of the monetary unit and its denomination.

Epigraphical style can be resorted to as a means to determine the approximate time of the coin, but this is feasible only in the few cases in which the stylistic distinctions can be established.

It would be very helpful if we knew the dates of establishment of the towns which cast spades for which specimens are known. Of the many mints which appear in the legends only those of Tung Chou (Eastern Chou) and one of An-yang can be dated. The determination of the dates of their establishment contributes much to the dating of the coins of these two cities, and the dates of their coins corroborate our hypothetical premises for the reconstruction of the evolutionary stages of the spade coinage.

There are some mint towns, for which the date of their establishment or of their coinage is unknown, whose geographical location reveals the approximate time when they cast the spades which bear their names. Hsiang-p'ing in southern Manchuria and Lin on the Yellow River in Shansi are towns of this kind. A study of their coinage, which will be dealt with when the Late Spade is treated, also helps confirm our chronology.

The traditional classification of the spades which has been followed until recently by all numismatists has been based on their shapes with such terms as "hollow-handle spade" or "spade money," "pointed-foot spade," "square-foot spade," and "round-foot spade." Chüng Chia-hsiang and Okutaira appear to have found this classification inadequate. Therefore, in their works they have included the shape of the coin's shoulders as an additional criterion for classification. Thus they formulated such terms as "pointed-(or raised) shoulder-pointed-foot spade" and so on. Basically, however, they still follow the old method. Some Western numismatists have given certain spades a specific name such as "weight money." In this case again, the primary consideration is the coin's shape.

To numismatists who are historically minded, the traditional classification is inadequate and should be revised. In the present study we propose a classification according to the coin's chronology. Thus we may divide the spades into four major groups: the Prototype Spade, the Hollow-handle Spade, the Old Spade, and the Late Spade.

Within each group we may further divide the coins according to their design. This aims not only at differentiating fine distinctions within a major group and their possible chronological sequence but also at investigating the regional character of a particular type. For instance, in the group of the Hollow-handle Spades there are two types differing radically in design. Although these two types appear to have been in circulation at the same time, their different designs suggest that they circulated in different areas. This is confirmed by the location of their mints as will be pointed out later.

However, some words of caution must be added. First, the coin specimens preserved and reported today may not represent every shape or every variety of coin that has existed, though we believe that they represent the majority. It is on the basis of preserved varieties that this study is made. Secondly, though money itself has an inherent tendency toward uniformity in design, local varieties and local discrepancies in the time of adoption of a new design must be regarded as inevitable. This is especially true in a period, such as that of Chou China, when customs in general and political and economic institutions in particular vary from area to area. The following presentation will be better evaluated and understood with these considerations in mind.

A. Prototype Spade

Specimens of this group of spades are illustrated on Plates IV—V. Their characteristic feature is a faithful resemblance to the original farming tool. Some of the coins are very large; some are a little smaller. Some possess all the practical features of the tool; some preserve them less. The criterion for their classification as one group is that their socket is like the tool's socket and extends onto the blade with the reinforcing ridge at the lower end of the socket remaining unchanged.

Most of the coins of this group bear no legend, though a few do. Judging from their size and design, the ones with a legend appear to be little later than the ones without. The paleographic style of the legend is the same as that of the early Chou bronze inscriptions.

The specimen in the collection of the American Numismatic Society is the smallest of all the specimens of the group known to us. It weighs 105.10 grams.

B. Hollow-handle Spade

Belonging to this group are two distinctively different types. One type ( Plates VI and VII, 1) has pointed shoulders and pointed feet. Hereafter it will be designated as H. H. Spade I. The other type ( Plates VII, 2; VIII—X) has flat shoulders and mildly curved feet. Hereafter it will be designated as H. H. Spade II. The differences in their design are so marked and their affiliation to the original tool is equally so close that it does not seem to be possible that one type is a variation or a later development of another. As has been mentioned before, they seem to be two varieties of spade coinage developed from two differently shaped spade tools. The existence of these differently shaped tools is evidenced by the actual specimens excavated in China and by the fact that there were two different names, ch'ien and pu, for spade money in the Chou literature. (See pp. 90—93).

As implied in the term which we have chosen to designate this group of spades, their characteristic feature is the hollow handle which is found on each of them. The difference between the hollow handle of this group and the socket of the Prototype Spade, and for that matter also of the original tool, lies in the fact that the handle stops at the coin's shoulders and does not extend onto the blade. Their further deviation from the shape of the tool marks their lateness in origin and serves as a convenient criterion for their classification.

This group has the design of three parallel vertical lines on both obverse and reverse. The central line, which is shorter than the side lines, resembles the reinforcing ridge in the center of the tool, and the two side lines are the vestiges of the sides of the socket on the blade.

Most of the coins of this group bear an inscription, though quite a few do not. This is another evidence that they are later than the Prototype Spade but earlier than the spades which always have a legend. The legend may be a numeral, a character from the ten "heavenly stems" or the twelve "earthly branches," or the name of a town or city, which is likely to be the mint. As far as we know, among the hollow-handle spades which have been reported, those of Mi ( Plate X, 2) are the first ones in which the legend has the monetary unit chin in addition to the mint name. On the whole the paleographic style of the legend of the Hollow-handle Spade appears to be the same as that of the Prototype Spade, namely, the "great seal character" style.

It is to be noted, however, that among the Hollow-handle Spades there are still a third and a fourth type, both of which seem to be later in origin than the two types described above. In design, the third type (Plates XI—XII, hereafter to be designated as H. H. Spade III) differs radically from H. H. Spade I and slightly from H. H. Spade II. Their principal distinctive feature is the fact that the three lines on their obverse and reverse are not parallel, but starting from the point where the handle joins the blade, the outer two lines diverge towards the tips of the feet. The Hollow-handle Spades of Tung-Chou (Eastern Chou), Lu-shih, Wu, Wu-an are commonly seen representatives of this type.

However, the designs of this third type are not homogeneous in all details. While the shoulders of the Eastern Chou spades are flat, those of the other three mints are slanting. While the outer of the three lines on these spades start from the center at the top of the blade and diverge to the tips of the feet, the three lines on the spades of other mints, such as "An -hsiang," run parallel and are symetrically placed. But their common features in design, size, and weight are such that a treatment of them as one type is warranted. All indications are that this type, i. e., H. H. Spade III, is probably a later development of H. H. Spade II. The insufficiency in the number of preserved specimens prevent us from any definite conclusions on this point.

The fourth group ( Plate XIII, 1—2, hereafter designated as H. H. Spade IV) are very few in number. The Encyclopedia of Old Coins contains illustrations of two specimens (Nos. 608 and 813), which are the only ones known to us. The authenticity of No. 813 seems reliable; it was in the possession of Fang Jo, an astute collector, who made and published a rubbing of the coin. Both specimens are small in size. Their blades are somewhat square in shape, plain on both sides, with no design whatsoever. Both the shoulders and the feet are flat. The date of their origin does not seem to be earlier than H. H. Spade III.

The following chart will summarize the descriptions of the four types of the Hollow-handle Spade:

Handle (socket) Stops at the shoulders Stops at the shoulders Stops at the shoulders Stops at the shoulders
Shoulders Pointed Flat Flat, slanting Flat
Foot or feet Two pointed feet Foot mildly curved Foot mildly curved Flat foot
Width of shoulders and foot Equal Equal Foot slightly broader Equal
Designs Three parallel vertical lines on both sides Three parallel vertical lines on both sides Central line vertical, sidelines diverging to tips of feet None
Size Large Large Small Very small
Legend Numeral, independent character, mint name Numeral, independent character, mint name Mint name Mint name

The period in which the Hollow-handle Spades circulated appears to have been considerable. Together with the Prototype Spades they covered the greater part of the time span of the entire spade coinage. An evidence of this is found in the Hollow-handle Spades bearing the name "Eastern Chou."

Eastern Chou was a feudatory which was established in the last year of King K'ao (426 B. C.) or in the first year of King Wei-lieh (425 B. C.).66 Thus the coins bearing the name of the feudatory must have been cast about 425 B. C. or shortly thereafter. At this time all feudal princes and the minor nobility had availed themselves of the privilege of coinage (see below, p.224). There is additional evidence for a late date for the Eastern Chou spades. First, their legend does not consist merely of a numeral or "serial mark" but includes also the name of the mint. Secondly, the three lines in their design are not parallel as on H. H. Spade II (which resembles the features of the original tool) but diverge toward the tips of the feet. This deviation indicates a late appearance. Thirdly, the Eastern Chou spade is one of the smallest of the Hollow-handle Spades. The measurements of the rubbing of Fang Jo's specimen67 are: total length, 70 mm.; shoulder width, 35 mm.; foot width, 37 mm.; blade length, 40 mm.

The largest specimen of the H. H. Spade II in the collection of the American Numismatic Society measures correspondingly 102, 53, 53, and 61 mm. This specimen is the older one not only because of its larger size but also because its legend consists solely of a numeral pa (eight). The Eastern Chou is considerably smaller in size and thus must be later, for the smaller a coin is, the later it is in date.68

End Notes
66 Shih-chi, IV, 32b—33a; Edouard Chavannes, Les Mümoires Historiques de Se-ma Ts'ien, Paris, 1895, 300 —301. Many historians have mistaken the eighth year of King Hsien of Chou (367 —6 B.C.) for the year of the creation of the Eastern Chou feudatory. Their mistake is caused by a misunderstanding of a passage in Shih-chi, XLIII, 17b, where it is recorded that the states of Han and Chaodivided Chou into two parts, in the eighth year of King Hsien. This event refers to the separation of the royal domain of Chou into two areas under their separate influence, and it does not refer to the creation of the Eastern Chou feudatory. As far as we know, Lü Tsu-ch'ien image (1137 — 1181) seems to be the only historian in the past who did not make this mistake (see his Ta shih chi, in the Chin-hua ts'ung-shu, I, 13b).
67 Yüeh-yü ku-huo tsa-yung (Coole, 290), the first specimen. Reproduced on Plate XII, 3.

C. The Old Spade

The outstanding characteristics of this group of spades are the absence of the hollow-handle (socket), which are present in the two previous groups, and the split of the foot into two square feet. The representative specimens of this group are the spades of An-i ( Plates XIV, 4—5; XV, 1), Liang ( Plate XIV, 1—3), Chin-yang ( Plate XIII, 3—5), etc. They are called "weight money" by Western numismatists. They still have a handle, but it is no longer hollow; it is flat and solid. Changes have taken place in the shape of the shoulders: some of them are round, and some are angular. Except for the "regular" spades of Liang, the spades of this group all have two feet which are uniformly square. Some of the spades have a central vertical line on the obverse and some do not. In the latter case the space is completely occupied by the legend. The reverse of some of them is plain, and some have a character, which represents an abbreviation of the mint name in the obverse legend or indicates the nature of the money, such as ch'ung (token).

The appearance of the Old Spade was accompanied by an increase in number of denominations. The monetary unit of the spade is the chin. The old spades of An-i and Chin-yang are each made up of three different denominational sizes: two chin, one chin and half chin. For the spades of other mints which are preserved today, we have only the two chin denomination (such as that of Ch'ui), or of the one chin and the half chin denominations (such as those of Yü), or of the one chin (such as that of Yüan), or of just the half chin (such as that of Lu-shih). Since, in both shape and design, the coinages of these mints belong to the same group as those of An-i and Chin-yang their complete denominational system must also have been the same. Otherwise the specification of "two chin," or "one chin" or "half chin" alone would be meaningless. The lack of all three denominations for these mints must be due to incompleteness in their preservation.

Mention must be made of the coinage of the Old Spade period in Liang. Besides its regular Old Spades of the type of An-i and Chin-yang, the mint cast also two special series of spades. One is designated as chüng or "standard," and the other is designated as ch'ung or "token" ( Plate XV, 3—5; XVI, 1). The shape and design of the "token" spades of Liang are exactly the same as those of its regular Old Spade and the Old Spades of other mints. Its "standard" spades vary slightly in shape, but their basic features are similar to the common features of the Old Spade. Both series seem to have been issued for interstate commerce. (See further discussion, pp. 137—143.)

As has been indicated, the legend of the Old Spade is made up of three items: the name of the mint, the monetary unit, and its denomination whenever the specification of the denomination is necessary. Only the old spade of Fün ( Plate XV, 2) varies from the rule. Its legend is fün pu, with fün being the mint name and pu the designation of the money. This is the first and the only case in which the monetary designation pu is found on a spade coin.

There is no positive evidence with which the date of the Old Spade can be determined. That it is later than the Hollow-handle Spade is unmistakable. Except for its general form, we do not find any of the significant features which mark the Hollow-handle Spade and which are reminiscent of the characteristics of the tool. As will be discussed below, the "Regular" and the "Token" spades of Liang (also known as Wei) appear to have been issued when Liang, at the peak of its power, dominated Chou China. This period is roughly between 425 B. C. and 344 B. C. and may well be the time when the Old Spade was in circulation.

Judging from its shape and design the Old Spade seems to have developed out of H. H. Spade II and III. The shape of their shoulders and feet which is, as a rule, flat and angular or square are similar. H. H. Spade I with its sharply pointed shoulders and pointed feet does not appear to be its predecessor.

At this point one may question what happened to the coinage of H. H. Spade I while H. H. Spade II and III developed into the Old Spade. It would seem that the shape of H. H. Spade I continued to be the shape of the coinage in the region where it had been current while H. H. Spade II and III were under further evolution in design. Its size, however, must have been reduced and its weight diminished to meet changing circumstances. Insufficient information prevents us from forming a definite conclusion.

End Notes
68 In Shan-chai chi-chin lu (Coole 346), Ch'uan lu, I, 54b —55a, Liu T'i-chih image the author reproduces a picture made from the rubbing of an extraordinarily large hollow-handle spade in his possession of the type of H. H. Spade III. In length it measures 164 mm., the width at its shoulders is 102 (including the two protruding points; actually only 88), the width at its foot is 102. So far this is the largest specimen reported of the late hollow-handle spade. Its legend reads "Lu-shih Niehchin" (dark metal money of Lu-shih). We have no way to determine its authenticity. If genuine, it would provide additional and important information on the monetary system of the day.

D. The Late Spade

The great majority of the spade coins preserved today belong to this group. They occupy the bulk of any collection, private or public, of Chou coins. According to shape and design they may be divided into four types:

Late Spade I—Pointed shoulders and pointed feet (Plates XVIII-XX; XXI, 1–6). Late Spade II—Square shoulders and square feet ( Plates XXI, 7; XXII-XXIV). Late Spade III— Round shoulders and round feet (Plates XXV; XXVI, 1–2). Late Spade IV—Round shoulders and round feet with three holes ( Plates XXVI, 3; XXVII).

A glance at the coins will convince us that Late Spade I developed out of H. H. Spade I, which have pointed shoulders and pointed feet. Late Spade II descended from the line of the Old Spade of which the predecessor is H. H. Spade II. The origin of the shape of Late Spade III is more difficult to decide. It appears very much to have developed out of the shape of the "Regular" spade of Liang, ( Plate XV, 3), both of which belong to the Old Spade (earlier than Late Spade III) and have a round handle, round shoulders and round feet. It might have been an imitation of Late Spade IV which was the design of the official currency of the state of Ch'in and will be discussed presently.

For each of the types there are two sizes. The large-sized specimens in the collection of the American Numismatic Society weigh approximately twice the smaller ones. Some small-sized specimens of Late Spade I and II have specification of their denomination in their legend, which appears either on the obverse together with the mint name or on their reverse (for examples see Plates XVIII, 2; XXII, 1). This is always the character pan for "half." The denomination of the larger size, which weighs twice as much as the small, must then be "one," referring to the monetary unit of the spade, the chin. Both the large and small-sized coins of Late Spade IV bear their denominational designations on the reverse (for examples see Plates XXVI, 3; XXVII). The denomination al designation of the large spade is liang (Chinese ounce), and of the small is "twelve chu" (chu being a very small unit of weight). Since at this time one liang was made up of twenty-four chu, the smaller coin is a "half" of the larger. Although different in size, the large and the small coins of each type have the same design. The characteristics of the types are shown in the chart on the opposite page.

An interesting phenomenon of the Late Spades is their lack of complete uniformity in design. Let us take the coins of Ta-yin as an example ( Plates XVIII, 1; XIX, 3). The spades of this mint are of the type of L. S. I, but a few have square shoulders instead of the predominating pointed shoulders. A few others have square shoulders and square feet. These exceptional shapes are obviously

L. S. I L. S. II L. S. III L. S. IV
Handle Square; two vertical lines converging on lower end Square, one vertical line Round, Plain Round, a hole
Shoulders Raised & pointed Square Round Round
Body One vertical line in center, legend Vertical line on handle continuing down through center, legend Plain legend Plain legend
Feet Pointed Square Round Round, a hole on each
Legend Mint name or combined with the denomination Mint name Mint name Mint name
L. S. I L. S. II L. S. III L. S. IV
Handle Square, one vertical central line Square, one vertical central line Round, plain Round, a hole, a numeral (serial mark?)
Shoulders Raised & pointed Flat and square Round Round
Body Two parallel vertical lines on the sides Central line on handle continuing down through body, two side lines starting from end of handle and diverging towards tips of feet Two lines starting from end of handle and diverging towards tips of feet Denominational specification
Feet Pointed Square Round Round, hole on each foot
Numeral mark or denomination Most have a numeral in center or on side Few have a numeral, placed the same way, some have a denomination A numeral in center A numeral on handle and a denominational specification on body
influenced by the shape of L. S. II. Although the forms of their shoulders and feet have changed, other features, however, remain.

Adoption by a mint of an alien coin type was one cause of the lack of uniformity in design. This is illustrated in the coinage of Chin-yang ( Plate XXI, 1,3). During the period of the Old Spade, Chin-yang cast the spade with square feet. There is every reason to expect that in the days of the Late Spade the mint would have cast Late Spade II, which developed from the Old Spade. Instead it cast the type of Late Spade I. Some other mints changed the shapes of their coinage once or twice within the same period. Thus we find Chung-tu cast both L. S. I and II. Towns such as Lin cast three types of the Late Spade.

The reasons for these changes in design may have been technical, economic, or political. During the Chan-kuo period in which the Late Spade circulated, the economic relations between different areas had become much closer. Active inter-regional trade leads naturally to constant exchange and assimilation of local customs and institutions. Coinage was likely to have been affected by this process of general cultural assimilation or borrowing. The replacemant of one type of coinage by another, or the borrowing by one mint of the type of coinage of another, may have been a result of the superior economic influence of the latter.

Changes in coin type for political reasons may be detected in the coinage of An-yang. The An-yang spades in question belong to Type IV ( Plate XXVII, 2). Spades of this type uniformly bear the monetary unit liang on the large size and its half (twelve chu) on the small. This type of the Late Spade differs in its monetary unit from the other three types, for which the unit is the chin. Since it has a special unit, it must have been a spade coinage of a special area, which was most likely the territory of the state of Ch'in. The round coin of Ch'in which was issued after Ch'in unified ancient China and unified the monetary system is designated liang. The identification of the same monetary unit on coinages of consecutive periods indicates that they circulated in the same area.

To be sure, students of Chinese history will find three towns by the name of An-yang during the Chou period. One is mentioned by Li Hsien and others in their notes to a passage in the Hou Han shu (Book of Later Han) and is located by them in present southeastern Shantung.69 A second, which was a town in the old Tai state in present northern Shansi, is recorded in the Shih-chi.70 A third is also recorded in the Shih-chi, where it is stated that in 257/6 B. C., "After the Ch'in armies conquered Ning-hsin-chung, it was renamed An-yang."71 This An-yangwas located southwest of the present city of like name in present northern Honan.

The first An-yang was located in the knife coinage area and cannot be the town which minted the spades. While both the second and the third are possibilities, we believe that the An-yang which cast the spades with round shoulders, round feet and three holes is the third An-yang, created by Ch'in in 257 B. C. We are led to this assumption by the fact that mints of this type of spade which have been deciphered and located were all situated close to the original territory of Ch'in. They are towns which were probably the first annexed by Ch'in in the course of its eastward expansion. It was probably after their conquest that Ch'in imposed on them its official coin type of round shoulders, round feet, with three holes.

Did Ch'in have an official type of spade coinage? The indications are that it did. In Shih-chi, VI, 50a, it is stated that in the second year of King Hui-wün (336/5 B. C.) the Ch'in "For the first time hsing money." In XV, 21a, in the chronological table for the Chan-kuo period it is stated under the same year, "The Son of Heaven (the Chou king) congratulated (Ch'in) for hsing money." Literally, hsing means "to put into usage or circulation." Scholars who argue for the late origin of metallic money in China have stressed these records beyond their proper limits. Their amplification of the significance of these data to be indicative also of other regions of ancient China is highly questionable, for we know that the spade coinage had begun as early as the early Chou. At that time the territory of Ch'in was the center of the Chou royal domain. The discovery of the hollow-handle spades in the area72 is an indisputable proof. It may be possible that the Ch'in people, who came from the west with a comparatively backward economic tradition, had discouraged the use of metallic money, but how far this held true of the actual situation is a question. The "putting into circulation the money" in Ch'in in 336 B. C. appears to be more like the establishment of an official currency with the implication of a state monopoly in coinage. This might have been the policy which heralded the unification and state monopoly of coinage on a national scale in 221 B. C. after Ch'in conquered the whole country. Judging from all the indications, this official type of currency seems to be that with round shoulders, round feet, with three holes. The establishment of an official type of currency was possible in a state which was as centralized as Ch'in was in 336 B. C. If this assumption and the assumption that the spade money of Ch'in was the Late Spade IV is correct, we may say the commencement of this particular type of Late Spade III was in 336 B. C.

The date for the Late Spade can also be gauged from the coinage of Lin of Type III, which mint had also cast Types I and II. In and before the fifth century B. C. the area in which Lin of a later date was located was still occupied by the Ti people, not by the Chinese. According to King Wu-ling of Chao (325—299 B. C.) to which Lin belonged, Lin was conquered and annexed by one of his ancestors.73This event must have taken place after the Chao state was officially created in 403 and before King Wu-ling became the ruler of the state in 325 B. C.74 That is to say that Lin as a mint of the Late Spade came into being only after 403 B. C. The commencement of its coinage may well be in the fourth century. In other words, the appearance of the Late Spade III of Lin cannot be earlier than 400 B. C.

As we know, between the termination of the spade coinage and the unification of the Chinese monetary system in 221 B. C. there was a period in which the round coin was the currency. In view of the fact that the round coins which are preserved or discovered are exceedingly few in comparison with the Late Spades, it seems that the period of the round coinage of Chou was very short. If we give twenty or thirty years for the period, we will have the year 250 B.C. as the approximate date for the end of the spade coinage. Thus, on the basis of all evidence available we get the following chronological order for the spade coinage:

Prototype Spade End of 12th or nth century B. C.
Hollow-handle Spade ca. 400 B. C.
Old Spade 400—340 B. C.
Late Spade 340—250 B. C.

This chronology reveals that the later the type of coin appeared, the shorter the period in which it was in circulation, and the faster it was replaced by newer types. The change of the types, or in other words, the adoption of new types, is accompanied by gradual reduction in size and weight and by the tendency to adopt more convenient forms of money.

End Notes
69 Hou-Han shu, 1643, XXCII, Part 2, 2b.
70 Shih-chi, XLIII, 27b—28a.
71 V, 34a.
72 Lo Chün-yü, Yung-lu jih-cha 11a.
73 Shih-chi, XLIII, 21a.
74 According to Shih-chi, XLIII, 19a, Ch'in seized Lin from the state of Chao in 328 B. C.

End Notes

64 In Shih-chi, XXXII, 3a—3b, Lü-shih ch'un-ch'iu, XV, 3b. In Shih-chi, IV, 12a and Shang shu (chapter of Wu-ch'üng, present text), III, 13a, the character ch'ien is replaced by ts'ai image meaning property, in general.
65 For the opinions of these scholars see Irito Seizō, Kōkogaku zasshi, XV (1925), 402 —412; Tsukamoto Yasushi image "Shina Kosen keijo no kigen ni tsuite," Kōkogaku zasshi, XV (1925), 491–499; Kuroda Kanichi, Kōkogaku zasshi, XVI (1926), 138, and "Shūdai kohei ko," Kōkogaku zasshi, XVII (1927), 670—677; Okutaira, op. cit. I, 7a and II, 84b; Chüng Chia-hsiang, "Shang-ku huo-pi t'ui-chiu," Ch'üan-pi (Chinese Numismatics), No. 4 (1941), p. 30—31 and 32—33; and No. 5 (1941)) p.22—24.


Both the provenance of the spades and the locations of their mints demonstrate that the spades were a coinage of a special region in ancient China.

In 1942 Chüng Chia-hsiang published a prefatory article to a study of inscriptions on ancient Chinese coins, in which he says that he has collected 244 different characters appearing on spade coins alone.75Judging from the average inscription on spade coins, at least eighty per cent of these, or about 195, must be names of mint towns. Many of the names of the mint towns, however, consist of two characters, and these mint names account for two thirds of the total. Thus the actual number of mint names would be no more than 140. We have ourselves collected 147 names of mint towns on spade coins in the collection at the American Numismatic Society, supplemented in some cases by those in coin catalogues published before 1939. Because of the difficulties mentioned in the introduction, not all of these names have been satisfactorily deciphered or located. The following table contains the names and locations of eighty-seven mint towns. The name of the mint is given in the first column of the table. In the second column is noted the name of the state to which the mint belonged or in which it was located during the Ch'un-ch'iu period (770—481 B. C.). In the third column is found the name of the state or names of states to which the mint belonged during, or in greater part of, the Chan-kuo period (403—221 B. C.). In cases where the name of a mint is found to be that of more than one town with simultaneous existence in one state or in different states, and the identification with a particular town is thus impossible, all the possibilities will be found noted in the table. (Table see p. 131–135)

The eighty-seven mint towns included in the above table should prove a sufficient number to show the geographical distribution of the towns which minted spade coins. In the Ch'un-ch'iu period, these mints were located in the royal domain of Chou and in the states of Yen, Chin, Chüng, Sung, Lu, Wei and other minor states. None was located in the states on the Shantung Peninsula (notably Ch'i, An-yang, etc), or in the present Hupeh Province, the original territory of the state of Ch'u.

The regional character of the distribution of the spade coins is further corroborated by their provenance. Hollow-handle Spades of various descriptions are reported to have been discovered in "Chung-chou" (Honan Province) by Li Tso-hsien, Pao K'ang, Ch'u Shang-lingand Ch'in Pao-tsan, in "Pien-chung" (Kai-füng) by Li Tso-hsien, in "Tsü-chou" (Chin-ch'üng in S. E. Shansi) by Pao K'ang, in "Lo" or "Lo-chung" (area around Lo-yang in N. Honan) by Fang Jo, in Müng-chin (N. Honan) by Ch'u Shang-ling, in "Kuan-chung"

Mint Location,770–481 B.C. Location, 403–221 B.C. Modern location Type of Spade cast by the mint
An-i image Chin Liang (Wei) S. W. Shansi O. S. (Old Spade)
An-yang image Chin? Chao N. Shansi L. S. (Late Spade) II
Wei? Ch'in (since 257 B. C.) N. Honan L. S. III
An-yin image Pro Chao N. Shansi O. S.
Ch'ang-tzŭ image Chin Chao, Han S. E. Shansi L. S. II
Chüng image Chüng Han C. Honan L. S. II
Ch'üng image Lu Ch'i (since 408?) W. Shantung H. H. S. (Hollow-handle Spade) II
Chai-yang image Liang N. Honan L. S. II
Ch'i image Chin Chao N. Shansi H. H. S. II; L. S. II
Chih image Chin Liang S.W. Shansi H. H. S. II
Chih image Chin Chao ? S. Shansi O. S.; L. S. II
Chin-yangimage Chin Liang S. W. Shansi O. S.
Chao C. Shansi L. S. I
Ching image Chüng Han N. Honan O. S.
Cho image Yen Yen N. Hopeh L. S. II
Chou image Chou E. Shensi H. H. S. II
Chou N. Honan
Chu image Chu Ch'u S. W.Shantung H. H. S. II
Han & Liang (361) E. Honan L. S. II
Ch'ui image Sung Liang E. Honan H.H.S.II; O.S.; L.S.II
Chung-tuimage Chin Chao C. Shansi L. S. II
Chung-yangimage Chin Chao or Liang S. W. Shansi L. S. II
Fün image Chin Chao or Liang S. Shansi O. S.; L. S. I
Chou or Ch'in or E. Shensi or L. S. II
Füngimage Sung Sung & Ch'u N. Kiangsu
Han-tan image Chin Chao S. Hopeh H. H. S. I; L. S. I
Hao image Ch'in Ch'in C. Shensi H. H. S. II; L. S. II
Hou image Chin Han or Liang N. Honan H. H. S. II
Huo (as alternative of image Hao) Chin Chao S. Hopeh
Chin or Chao or S. Shansi or L. S. I
Huoimage Chou Han C. Honan
Hsi-tu image Chin? Chao C. Shansi L. S. II
Hsia-pi-yang image Han? C. Honan L. S. III
Hsiang image Chou or Liang or N.W. Honan H. H. S. II
Chüng? Liang ? E. Honan
Hsiang-p'ing image Yen Yen S. Liao-ning L. S. II
Hsiang-yüan image Chin Liang S. E. Shansi L. S. II
Hsin (5) image All in spade coin area L. S. II
Hsin-ch'üng (7) image All in spade S. E. Shansi L. S. I; L. S. II
Jang-yin image Chin Han? S. E. Shansi L. S. II
Kao-tu image Chin or Liang or N. Honan L. S. II
Chüng Han N. W. Honan
image N. W. Shantung H. H. S. II
Kuai image Kuai & Chin Han N. W. Honan L. S. I; L. S. II
Kung image Wei Liang N. Honan O. S.
Kuo image Sung Ch'u E. Honan H. H. S. II; L. S. II
Lai image Sung Ch'u E. Honan H. H. S. II; L. S. II
Li-shih image Chao W. Shansi L. S. II; L. S. III
Liang image Liang N. E. Honan O. S.; L. S. II
Lin image Chao W. Shansi L. S. I; L. S. II; L. S. III
Liu image Sung Ch'u N. Kiangsu H. H. S. II
Lu image Chin Han S. E. Shansi L. S. II
Lu-shih image Han W. Honan H. H. S. III; O. S.; O. S. II
Lu-yang image Ch'u C. Honan H. H. S. III; O. S.
image Chin Chao S. Shansi H. H. S. I
Mi image Lu S. W. Shantung H. H. S. II
Nieh image Chin Liang S. E. Shansi O. S.; L. S. II
Ning image Liang N. Honan O. S.
Pei-ch'iu image Ch'i N. W. Shantung L. S. II
Pei-ch'ü image Chin Liang S. W. Shansi L. S. II
Pei-tzŭ image Chin Chao S. Chansi L. S. I
Pi-yin image Han C. Honan O. S. (?)
P'i-shih image Chin Liang S. W. Shansi L. S. I
P'ing-chou image Liang C. Shansi L. S. II
P'ing-yang image Wei or Liang? or N. Honan or L. S. II
Chin or Han (or Chao or Liang) or S. W. Shan si or
Lu S. W. Shan tung
P'ing-yin image Chou N. Honan L. S. II
P'ing-yüan image Chin Chao N. W. Shan tung L. S. II
Po image Sung Ch'u? E. Honan H. H. S. II
P'u-pan image Chin Liang S. W. Shansi O. S.
P'u-tzŭ image Chin Liang W. Shansi L. S. II
Shan-yang image Liang N. Honan O. S.
Shang-ch'iu image Sung Sung? E. Honan L. S. II
Shang-piyang image Han? C. Honan L. S. III
Sung image Sung Sung E. Honan H. H. S. II
Ta-yinimage Liang or Han N. W. Honan L. S. I; L. S. II
Tu image Chou Ch'in W. Shensi H. H. S. II
Tu-yang image Chou Ch'in W. Shensi L. S. II
T'un-liu image (2) Chin Han S. W. Shansi L. S. II
Tung-chou image Tung-chou N. Honan H. H. S. III
T,ung-t'i image Chin Liang ? S. Shansi L. S. II
Tzŭ-shih image Chin Chao C. Shansi L. S. I; L. S. II
Wün-yang image Lu Lu C. Shantung L. S. II
Wu image Chüng or Han or N. W. Honan L. S. II
Wu image Chin Chao Liang, Ch'in (after 396) or C. Shansi E. Shensi H. H. S. III
Wu-an image Chin Chao N. Honan H. H. S. III; L. S. II
Wu-p'ing image Chin Chao N. Honan L. S. II
Yang image Han N.W. Honan H. H. S. II; L. S. I
Yang-ch'üng image Han N. Honan L. S. II
Yang-i image Chin Chao C. Shansi L. S. II
Yang-jün image Chüng Han C. Honan L. S. II
Yin-chin image Chin Liang, E. Shensi O. S.
image Chin? Ch'in (after 332) Han? N.W. Honan H.H. S. II
image Chin Liang S. W. Shansi ;H. H. S. II; O. S.
Yü-yang image (since 665 B. C.) Yen Yen N. Hopeh L. S. II
Yüanimage Liang S. W. Shansi O. S.
(C. Shensi) by Lo Chün-yü.76 The old Spades of An-i are reported to have been discovered in "Chin-chung" (Shansi), and the various Late Spades are said to have been recovered in "Chin-chung" (Shansi) and "Shan-yu" (Shansi), in "Chi-hsing" (Hopeh) or Chih-li (Hopeh) by most of the numismatists mentioned above.77 The most recent discovery of a number of H. H. Spade I in the Chou tombs by Kuo Pao-chün was made in Chi county, northern Honan. In the Ch'un-ch'iu period, these areas were the territories of the royal domain of Chou and of the states of Chin, Ch'in, Yen, Wei, Chüngand Sung, in all of which mint towns of spade coins were located.78

If we examine further the locations of the mints of the various types of the spades, we will find that within this extensive area, certain types of spades are found to be local varieties of particular regions. H. H. Spade I and Late Spade I, which developed out of the former, appear to be a type of the region between the Yellow and the Fün Rivers in what is now southern Hopeh and central Shansi. H. H. Spade II was the type along the Wei g, in central Shensi, the Yellow River in Honanand the Chi River in southwestern Shantung. H. H. Spade III was the type in the region of present western Honan, and Late Spade IV was the type of the state of Ch'in of the late Chan-kuo period, the territory of which corresponded to present western Shansi, western Honanand Shensi.

Some of the towns originally belonging to Sung, Lu and other states were annexed by Ch'u during the Chan-kuo period, and some originally belonging to Lu and Chin were annexed by Ch'i. Both Ch'i and Ch'u had their own special coinages, the knife and Yüan Chin respectively. Coins show that change of political domination did not bring about abolition of the previous coinage and the exclusive adoption of that of their new overlords by these towns. It is true, of course, that a gradual change to wider circulation of the conquerors' currency took place.

Although a specific type of coin, as the spade or knife, was regional in the location of its mints and principal circulation, it must be understood that inevitable trade between the various areas would cause limited amounts of currency to be used in outside areas.

End Notes

75 Chüng Chia-hsiang, "Ku huo wün-tzŭ hui-pien tzŭ-hsü" (Preface to a collection of inscription characters on old coins), Ch'üan-pi, No. 11, (1942), 42.
76 See Li Tso-Hsien, Ku ch'üan hui (Coole, No. 266), Part I, X, 1b; Li Tso-hsienand Pao K'ang, Hsü ch'üan hui (Coole, No. 199), Part I, II, 7b; Ch'u Shang-ling, Chi-chin so-chien lu (Coole, No. 9), "fan-li" (Rules for the compilation) and II, 8a—8b; Pao K'ang, Kuan-ku-kü ch'üan-shuo (Coole, No. 296), 1a; Pao K'ang ed., Hsü ch'üan-shuo (Coole, No. 202), 4a; Fang Jo, Yüeh-yü ku-huo tsa-yung (Coole, No. 290), under the "Han-yang" hollow-handle spade; Lo Chün-yü, Yung-lu jih-cha (Coole, No. 392), 1934, photostat ed., 10b; Chüng Chia-hsiang, "Shang-ku huo-pi t'ui-chiu" (Researches in ancient coins), Ch'üan-pi, No. 9 (1941), 23.
77 See NiMo, Ku-chin ch'ien lüeh (Coole, No. 245), II, 11b. Ch'u Shang-ling, op. cit., I, 4a, 14b, 16a, 16b, 18a; II, 4b; Li Tso-hsien, op. cit., Preface III, 12b; Kuan-ku-kü ch'üan-shuo, 1b; Hsü ch'üan-shuo, 1b; Pao.K'ang, Kuan-ku-kü ts'ung-k'ao, III (Coole, No. 299), 7b; Ch'üan-pi, No. 5 (1941), 23.
78 A nineteenth century collector named Chin Hsi-ch'ang reports that during the reign of Emperor Ch'ien-lung (1736—1795)" several tens to a hundred" of hollow-handle spades bearing legends of "Wu" and "Lu-shih" were unearthed in mountains of Yü-hang in Chekiang Province (Ch'ing-yün-kuan shou-ts'ang ku-ch'üan shu-chi, Coole, 101, IX, 4b—5b). The locations of Wu and Lu-shih are indicated in the table. Yü-hang is in northern Chekiang near the sea coast. During the Ch'un-ch'iu period it was in the territory of Yüeh. It is not known how old the remains were in which the coins were discovered, therefore it is difficult to determine when those spades were brought there. They may have been brought at the time when they were still in Circulation in the interior of ancient China.


( Plates XV, 3—5; XVI, 1)

The term, "Special Old Spades." like other terms used to classify spade coins, is given purely for convenience's sake. With it we refer to a large group of spades of Liang which differ from its regular series of Old Spades. Both groups appear to have circulated at the same time. The legends on the special group are the longest found on any spade coins, whatever their type, and their content is also unusual. Inasmuch as the group differs from others substantially in monetary significance, they deserve special discussion.

These spades may be divided into four types, The legends of two of the types are made up of six characters, while the legends of the other two have seven and eight respectively. The six-character and the seven-character legends are not very difficult to read, but the eight-character legend has been, to use Lacouperie's words, "a stumbling block" to numismatists. To be sure, quite a few different readings for it have been suggested,79 but each of them is based on unacceptable conjecture. It is to Tung Yu-ch'üig (1791—1821) and Kuo Mo-jo (1892—) that we must give credit for its correct reading.80The four legends are as follows:

  • Liang chüng Shang chin tang lieh image or "Liang standard superior money equal to one lieh." Plate XV, 3.
  • Liang pan Shang erh chin tang lieh image or "Liang superior money, half (unit), two equal to one lieh." Plate XV, 4.
  • Liang ch'ung chin chin tang lieh image or "Liang money to be used as one chin and equal to one lieh." Plate XV, 5.
  • Liang ck'ung chin wu tang lieh shih erh image or "Liang money to be used as five chin and equal to twelve lieh." Plate XVI, 1.

The weights of the specimens in the cabinet of the Museum of the American Numismatic Society are shown in the following chart:

Type Average Weight Weight of Heaviest Specimen Weight of Lightest Specimen
a 12.61 (average of 17 specimens) 16.00 10.80
b approximately half that of c81
c 12.18 (average of 8 specimens) 15.05 7.21
d 23.54 (average of 7 specimens) 28.02 17.40

Liang, the first character in each of the legends, is the name of the mint or issuing city or authority, as are the first characters or initial combinations of characters in the legends of all coins of the Chou period.

Chüng, the second character in the legend of a) means "regular" and "standard." Some numismatists understand it as meaning "whole" or "a whole one," as against the "half" in the legend b).82 This, however, does not seem to be a plausible explanation.

Pan, the second character in legend b) means "half," a denominational term used for all types of spades except the Hollow-handles. It must be used here in reference to the denomination of one of the other three types of the Special Old Spades of Liang. Since coins of legends c) and d) are of one group distinguished by the monetary designation "ch'ung" in their legends and are different from type b), the type with which type b) with the "half" denomination is related must be type a), the full "unit" spade of Liang.

The size of the spade of the "half" denomination is much smaller than that of the "unit" spade (see Plate XV, 3—4) with legend a) and legend b). According to Kuo Mo-jo, the weight of the spade with legend b) is approximately half of the spade with legend c). As shown in the above chart, the spade with legend c) weighs approximately the same as the spade with legend a). Therefore, the weight of the spade of type b) must be about half the weight of the spade of type a). In other words, the spade of type a) represents the coin of a whole unit, while the spade of type b) represents that of a half unit.

As specified in their legends, the monetary unit of the spades with legends c) and d) is the chin. Since the spades with legends a) and b) belong to the same group, the monetary unit of the latter must be the chin. The average weight of the spade with legend a) shown in the above chart is roughly the same as the weight of ordinary Old Spade of one chin denomination, about 13 grams.

The reading of the second character in legends c) and d) as ch'ung has been accepted by most numismatists and some epigraphers.83 The primary meaning of the character ch'ung is "to fill up," from which the meaning of "to be used as" or "to be reckoned as" is later derived. In the case of the spade with legend d), whose weight does not correspond to its denomination, ch'ung actually signifies what we mean by the word "token." It is most likely that, though specified as of one chin denomination, the ch'ung spade with legend c) weighed less than the spade of the "standard" type of the same denomination.

The last character in all of the legends is lieh, which has been erroneously identified and read as yüan by most numismatists. This will be further explained on pp. 207—211.

Anyone who reads the legends of the Special Old Spades of Liang cannot help being impressed and fascinated by the phrases "standard superior money," "money to be used as" and their equations to or exchange rate with the lieh. These expressions or specifications are not found either in the legends of any other Old Spades or of the spades of other types. Their unique nature leads us to believe that the Special Old Spades of Liang were cast to cope with new and special monetary situations.

Let us examine the history of Liang to see whether we may find the special conditions which may have called forth the type. In the early years of the Chan-kuo period in which Old Spades appeared there were three places bearing the name "Liang." One, designated either as "Liang" or "Shao Liang" (Small Liang), was located on the west bank of the Yellow River in the central part of eastern Shensi. Another, designated as "Nan Liang" (Southern Liang), was located south of the Ju River in central Honanand appeared to have been annexed by the state of Han about 376 B. C. Both towns were situated in an area far from the economic and political center of China of that day, and both were economically rather insignificant. Neither seems to have been the one which established the new and complex coinage under discussion.

The third city with the name of Liangwas Ta Liang or "Great Liang," capital of a state which was located on the eastern plain of ancient China close to the "Cross-road," and on the Great Canal waterway system connecting the Yellow River valley and the valleys of the Huai and the Yangtze Rivers. It was strategically situated in the state of Liang, also known as Wei (to be distinguished from the Wei state whose name is written differently though pronounced the same, and which was created many centuries earlier). In 362 B.C. the state abandoned its old capital of An-i in the west, moved to Great Liang, which was made its new capital.84 Before 340 B. C., the year Liang yielded to the state of Ch'in all its land and towns west of the Yellow River at its southern bend, the territory of Liang was extensive and well placed, cutting across the central part of ancient China and connecting Ch'in in the west (modern Shensi) and Ch'i in the east (Shantung Peninsula). In the north it had possession of the old Chung-shan state in central Hopeh, and in the south it may have penetrated deep into the valley of the Huai River. In other words, the territory of the state occupied horizontally the central part of Chou China in the Chan-kuo period. In this stratigically situated territory traversed the Yellow River, the Fün River, the Chang River, the Southern Chi River, and the canal system of the Great Canal. All of these were important waterways in ancient China and constituted a communication system to and from all directions.

Upon such a promising natural potentiality were exerted the efforts of a wise administration. The early history of the state is filled with laudatory stories about the administrative activities of the best statesmen of the age. During the reigns of Marquis Wün (425 —397 B. C.), Marquis Wu (396 —371 B. C.) and King Hui (370 — 319 B. C.) the state underwent fundamental reforms in both its politics and economy. It abandoned the traditional aristocratic rule by blood, and redistributed the land (at least in part of the state), aiming at bettering the life of the peasants. Ways and means were worked out to promote production in agriculture, and promotion of commerce had been encouraged. As a result, as Prof. Ch'ien Mu has rightly pointed out, the state acheived a sort of hegemony to be emulated by other feudatories over a period of some eighty years from 425 to 344 B. C.85 It was during this Old Spade period that the "Special Old Spades of Liang" were in circulation. Indications are that Liang cast these special spades in addition to its ordinary Old Spades, after the state moved its capital to Great Liang in 362 B. C.

At this time commerce in China was expanding. With their state the most powerful at the time, situated in a key economic area through which merchandise to and from all parts of China must traverse, the rulers of Liang may have found it necessary and advantageous to issue special coins to facilitate the business transactions between their merchants and those of other regions. This motivation is clearly manifested in the fact that every one of the Special Old Spades was given not only a denomination in terms of the local monetary unit chin, but also a denominational equivalence in terms of the lieh. The lieh seems to have been the weight unit of the early spade coinage and remained to be such in some parts of the spade area after Old Spades had made their appearance. The ever present specification of the equation between the chin and the lieh on these Special Old Spades indicates that they were cast as a sort of interregional currency between the area where the chin was the monetary unit and the area which used the lieh.

End Notes

79 For the various readings of the legend see Ku ch'ien ta-tz'ŭ-tien, VIII, 20a—21b; Ch'üan-pi, No. 24, p. 11; No. 25, p. 2; No. 26, p. 8.
80 The article of Tung Yu-ch'üng image is quoted in the Ku ch'ien| ta-tz'ŭ-tien, XII, 496a. Kuo Mo-jo's reading is in Liang Chou chin-wün-tz' ŭ ta-hsi k'ao-shih, 13b. It must be noted that both scholars arrived at the reading by calculating the weight of the spade with this legend, but the method used by Tung Yü-ch'üng is wrong, because he mistakenly understood the chin image(monetary unit of the spade) as exactly identical with the later weight measure chin image, which is made up of sixteen liang. Actually, during the Chou period the chin as a monetary unit weighed approximately as much as the liang. Therefore his reading is just accidentally correct.
81 The weight of type b) is according to Kuo Mo-jo, who seems to have weighed a specimen of this type.
82 Chüng Chia-hsiang, for example, holds this opinion; see Ch'üan-pi, No. 25, p. 9.
83 Other decipherments for the character are ch'i image meaning "uncommon" or "special" (Fang Jo, op. cit., see above, note 76; decipherment found under the illustration of a specimen of the Special Old Spade of Liang), hsin image meaning "new" (Ch'üan-pi, No. 25, p. 2) and k'ua image meaning "big" (op. cit., No. 26, p. 8). Tentatively we suggest hsüan image meaning "dark."
84 This date (362 B. C.) about a quarter of a century earlier than the traditional date, is arrived at from a statement in the old edition of the Bamboo Annals (See Ch'ien Mu, Hsien-Ch'in chu-tzŭ hsi-nien k'ao-pien, 135—142). Because its capital from this year on was Ta Liang, the state was thereupon called Liang.
85 Ch'ien Mu, op. cit., 126 and the section of tables, pp. 90 —91.



The problem of the origin of the knife coinage is much simpler than that of the spade. The character for tao, the name of the knife coinage, is the same as for the instrument from which it developed. In shape, the coin faithfully resembles the original implement, and there is no possibility whatsoever for the coin to be taken as an imitation of any other object.

In one of the first excavations in Yin-hsü, the last capital of the Shang dynasty, a bronze knife was recovered ( Plate XXVIII, 1).1 Its main features are identical with those of the early knife coin. Its body is slightly bent as is the Early Knife, and at the end of its handle there is a ring, which is also found on the coin. Its measurements, made from Li Chi's illustration, are 220 mm. in length and 25 mm. in width. Later two other bronze knives of the same shape were discovered at the same site.2 One of them had a richly decorated handle on which the ring was replaced with a "horse's head." It is 288 mm. long and 38 mm. wide at its broadest point. The other knife was plain, 228 mm. in length and 36 in width, with the usual ring at the end of its handle. Different from the earlier one, the handles of these two are slit from end to end. The purpose for which the groove was made is not known.

The American Numismatic Society has in its collection a knife ( Plate XXVIII, 2) which is said to have been discovered in Wei Hsien (Wei County) in eastern Shantung. It was badly oxidized through the ages, and in appearance gives the impression of being very old. Like the others described above, it has also a mildly bent body, but, unlike them, its blade is much narrower, being only 12 mm. at its broadest point, which is always the part of the blade which joins the handle. Its length is 220 mm. The ring at the lower end of the handle is exceedingly large. The handle which appears round in shape has a few parallel grooves, with four long ones running from end to end. Those grooves were obviously made to give the user a firm hold. This design on the tool reminds us of the two parallel lines on the handle of almost all of the knife coins, early or late. The two-line design on the handle of the coins certainly is in imitation of these grooves on the actual tool. Though this knife may not be the very type from which the first knife coins developed, it certainly suggests the features of the knife in general. The coincidence of the grooves on the tool and the raised two-line design on the coins becomes more significant when we consider that the reported location where the knife was discovered is in the area where the Early Knife coinage circulated.

Some numismatists believe that the knife after which the coin was modelled was a household implement, and some contend that it was a weapon. Li Chi regards it as the hsüeh mentioned in the K'ao-kung chi3 which is now incorporated in the text of the Chou li. The ancient Chinese hsüeh was something functionally like a whittling knife used to cut off thin slices from a piece of wood or bamboo. The shape of the knife specimens described above suggests such a functional use. In the Chou li it is stated that the length of the hsüeh is one ch'ih (foot) and the width is one tenth. The longest of the foot measures of the Chou period which have been discovered and reported is 225 mm., and the shortest is 219 mm. The measurements of the knives mentioned above correspond fairly well to those for the hsüeh recorded in the Chou li. This is additional support for Li Chi's identification with the hsüeh.

Which type of the knife coins preserved today is the "early knife ?" This is still a topic of controversy. In our opinion the early knife coins are the large knives of Ch'i, Chi-mo, An-yang and T'an ( Plates XXIXXXXVIII, 1), which is also the opinion generally held by most numismatists, save for Okutaira Masahiro and Chüng Chia-hsiang.

Chüng Chia-hsiang contends that the earliest knife coins are the "sharp-pointed knives."4 So also does Okutaira,5 though he is silent on the reasons for his belief. Chüng Chia-hsiangexpresses the following reasons for his contention: 1) Sharp-pointed Knives have a (thin) blade (in other words, they resemble more nearly the actual tool); 2) their legend is not a mint name; 3) the style of their inscriptions is mostly that of the "great seal character," while that of the large knives of Ch'i and the other three states is mostly in the style of the "small seal character."6

To his first point we may counter that not only is the blade of the Sharp-pointed Knives thin, but every part of them is thin. In fact, their handle ring is so thin and flat that, contrary to Chüng's supposition, it only remotely resembles that of the original tool. The thinness and fragile appearance of the knives suggest rather that they are late in origin. Concerning the style of inscription, Chüng Chia-hsiang's argument is not borne out by facts. Comparison of the inscriptions on the Sharp-pointed Knives and on the large knives of Ch'i and the other three states does not show marked differences in style. There are also archaic pictograms in the inscriptions of the large knives. Furthermore, it is his opinion that the Sharp-pointed Knives lasted a long time and remained in circulation as late as the Chan-kuo period.7 In other words, they are not only anterior to the Ch'i knives, but also contemporary with and posterior to the latter for certain periods. It is difficult to see how Chüng Chia-hsiang could reconcile the late date of these knives with their supposedly more archaic style of inscription. The fact is that the stylistic distinction alleged by him does not exist.

Of the three reasons expressed by Chüng Chia-hsiang the second is the most plausible, but here too he failed to make his argument convincing. Of the many hundred Sharp-pointed Knives preserved today only one specimen, that of Lin, or possibly two (if that of Liao reported by him be included), bears a mint name. Yet Chüng Chia-hsiang has admitted that as late as the Chan-kuo period, or the end of the Chou dynasty, Sharp-pointed Knives continued in circulation. Many of the Sharp-pointed Knives in circulation then must be without inscription of the mint name. This fact demonstrates clearly that the inscription of mint name alone, or rather the absence of mint name as the coin's legend, cannot be sufficient evidence for determination of its date. Among the Small Knives of the third century B. C. the great majority have no legend whatsoever. Can we say that those uninscribed late knives were earlier than the inscribed Ch'i knives? What makes one most skeptical about Chüng Chia-hsiang'salleged early origin of the Sharp-pointed Knives is the unbelievably long duration in circulation which he ascribes to them. Chüng Chia-hsiangcontends that they originated at the beginning of the Chou dynasty and lasted into the end of the period, covering about eight centuries. This would require that in such a long period no noticeable changes occurred to the coins with regard to their shape, length or weight, while in immediately adjacent areas the spade coin was undergoing constant change in those respects.

We believe that the knife coinage like the spade coinage had undergone many and marked changes from its beginning to its end, and we also believe that the Sharp-pointed Knives were considerably late in origin. In addition to their small size and low weight, the evidence in support of our belief is: First, Sharp-pointed Knives are reported to have been discovered together with Ming knives8 which, as admitted by all, were coins of the Chan-kuo period and consequently late in origin; secondly, the town of Lin had cast this type of knife.9 Lin was not established before 430 B. C. (see above p. 128). Since the Lin sharp-pointed knife is typical in every respect, the Sharp-pointed type of knife coinage could not have begun much earlier than this date.

The evidence for an earlier date for the large knives of Ch'i is to be found in their legend, which reads Ch'i tsao-pang ch'ang fa-huo image or "Everlasting legal money of Ch'i at the establishment of the State" (Plate XXIX).10 Here the phrase tsao-pang (establishment of the state) is the point of crucial importance.

When was the state of Ch'i established ? The answer varies with different numismatists. Some of them, and many historians too, follow the traditional account of the history of Ch'i and believe that it was established as a feudatory state after Lü Wang, popularly known as T'ai-kung Wang, the most important assistant of King Wün and King Wu of Chou, was enfeoffed with Ying-ch'iu in the territory of Ch'i after King Wu conquered the Shang dynasty in 1122 B. C. (conventional date).

Other numismatists are skeptical about such an early date for the coins although they still believe in the traditional date of 1122 B. C. for the enfeudation of Lü Wang by King Wu. To reconcile the conflict between their belief of the late origin of the knife coins of Ch'i and that of the traditional origin of the state, they advance a different interpretation of the phrase tsao-pang, establishment of the state. They argue that the phrase does not refer to the first creation of Ch'i in the twelfth century but to the hegemony the state attained during the reign of Duke Huan (685 —643 B. C.). They then naturally assume that the tsao-pang knives were cast in the seventh century. Of this interpretation Chüng Chia-hsiang is the exponent.11

Some other numismatists find the explanation of this school untenable, for the reason that, though Duke Huan achieved hegemony over other feudal states and even the royal court, he did not "establish" a new state. However, they share the conviction that the Ch'i large knives were late in origin. Happily, they find an event in the history of the state of Ch'i which could be interpreted in their favor, the usurpation of Ch'i of the house of Chiang (family name of Lü Wang and his descendants) by the house of T'ien in 386 B. C.12 The T'ien were a powerful noble family in the state of Ch'i from the thirties of the fifth century on. In 391 B. C. T'ien Ho expelled Duke K'ang of Ch'i to an island off the Shantung Peninsula. In 386 B. C. he established himself as the ruler of the state. In the opinion of these numismatists, the usurpation of Ch'i of Chiang by the T'ien is the "establishment of the state" referred to by the legend of the tsao-pang knives. The numismatist who strongly advocates this theory is Okutaira Masahiro.13

Most numismatists follow the first of the three hypotheses, few the second, and still fewer follow the third. After examining all relevant facts, however, we find none of the three satisfactory.

The third theory advocated by Okutaira does not fit into the chronology of ancient Chinese coinage. The large size, heavy weight, and the archaic epigraphical style of the coin legends preclude the possibility of the Ch'i large knives being coins of the fourth century. Though the power of the state of Ch'i had been transferred from one family to another, a new state was not created. In its institutions, legally, politically and economically, the old Ch'i continued in every respect. Even the very name of Ch'i was still kept by the usurper.

The second interpretation as advocated by Chüng Chia-hsiang has similar weaknesses. Duke Huan inherited the Ch'i created by his forefathers. Though he expanded the territory of the state and made it powerful, he did not establish a new one. For Duke Huan to have regarded the Ch'i during his reign as his own new creation would have run counter to the old Chinese political and ethical philosophy that the state was sacred property handed down by their ancestors to a ruler as a household was to an ordinary man. According to this philosophy the very existence of posterity was a favor from the ancestors and the survival of the state depended on their protection. Posterity owned nothing and created nothing, though posterity could glorify the state by aggrandization.

On the whole we agree with the first interpretation that the knife coins of Ch'i were cast when the state was first established, but we disagree with the traditional date of the state's establishment.

King Wu did not grant the territory of Ch'i to Lü Wang as his fief. He could not, for during his reign the territory which later came to be known as Ch'i had not been conquered by Chou and was still in the hands of the Shang people or their vassals. This has been proved by Prof. Fu Ssŭ-nien.14 Prof. Fu also contends that the very name of the supposed recipient, Lü Wang or Wang of Lü, suggests that the original fief of Wang was Lü and not Ch'i. Even Wang's son still bore this name, Lü.15 As we know, in ancient China after a noble was granted a fief or appointed to an office he was thereafter generally called by the name of the fief or that of his office. Therefore the Lü of Wang whose original family name was Chiang was undoubtedly the name of his fief or his main fief (see below). Lü as a feudatory in the early days of Chou was located west of present Nan-yang in Honan.

However, it is an undeniable historical fact that Ch'i was a state of Lü Wang's descendents. In the Ku-ming chapter of the Shang-shu, Lü Chi, Lü Wang's son or one of his sons, is already called by the title of "Ch'i Hou" or "Marquis of Ch'i" in 1079 B. C. There is no doubt that at this time the territory of Ch'i had been granted to Lü Chi. This is thirty-two years after that district was conquered by Chou, probably in 1111 B. C., according to the traditional date. The first recipient of the Ch'i fief might have been Lü Wang himself, for in the early historical literature he is depicted as a man who had lived for a considerable length of time. Ssŭ-ma Ch'ien records that he lived for "more than one hundred years."16 The old text of the Bamboo Annals states that he died in the sixth year of King K'ang, which falls in 1073 B. C. in the traditional chronology.17 If the latter record is reliable, it is possible that Lü Wang was the recipient of the fief.

Does the possible grant of Ch'i to Lü Wang mean the establishment of Ch'i as a feudal state ? It may or may not. A passage in the T'an-kung chapter of the Li chi states that "Since T'ai-kung (Wang) was enfeoffed with Ying-ch'iu (as his fief) down to the fifth generation his descendants always returned their dead to Chou for burial."18 This statement makes it clear that as late as the fifth generation Ch'i was not regarded as their home by the very noble beneficiaries. Our impression is that after the area of Ch'i was conquered by Chou in 1111 B. C. it might have been given to Lü Wang as an additional fief if he had lived that long, or to his son Lü Chi as his fief, with his fathers original fief, Lü, retained in the hands of other members of the noble family. If the recipient was Lü Wang, there is no reason to believe that he had resided there. As indicated in a passage in the Ku-ming chapter of the Shang-shu, even his son Lü Chi, who bore the official title of "Marquis of Ch'i," stayed at the Chou court and served as an important minister. If the recipient was Lü Chi, he was an absentee feudal lord. In neither case does Ch'i seem to have been established as a state, though the noble family might have received revenues from their fief.

It was the tradition of the Chou that the establishment of a feudatory must be accompanied with the completion of the following steps: 1) investiture including the casting of memorial bronzes; 2) the construction of the she, altar for the god of the earth and symbol of the existence of the state; 3) the erection of the ancestral temple of the ruling family; and 4) the construction of a walled capital. The completion of these requirements could be prolonged for some time. Since down to the fifth generation the nobles of Ch'i still sent their dead back to Chou for burial, there is more reason to doubt than to believe that Ch'i had been established as a feudal state before the fifth generation of the noble house.

This conjecture fits into Ssŭ-ma Ch'ien's account of the early history of Ch'i in Shih-chi, XXXII. During the first five generations the history of the noble family of Ch'i was full of internal feuds. Duke Hsien was the first to establish his capital at Lin-tzŭ, which remained such until its conquest by Ch'in in 221 B. C. Taking Lü Wang as the first ancestor, Duke Hsien was the fifth generation in Ch'i pedigree. It is quite possible that Duke Hsien may have been the one who "established" the Ch'i state, and who, if our foregoing hypothesis be correct, cast the tsao-pang knives of Ch'i. Duke Hsien ruled Ch'i during the reign of King I of Chou, which corresponds to 894—879 B. C. according to the traditional date. This assumption, while reasonable, cannot, however, be substantiated otherwise. Therefore, we may tentatively conclude that the tsao-pang knives of Ch'i may be as early as around 1079 B. C. and as late as the first half of the ninth century.

Is the tsao-pang knife, the earliest of all Ch'i knives, also the earliest of the knife coinage ? This is another question which deserves serious consideration. In addition to those of Ch'i, there are also the large knives of Chi-mo, An-yang and T'an ( Plates XXXIV—XXXVIII, 1.) Chi-mo, An-yang and T'an were all old states on the Shantung Peninsula in ancient China. Because they were later annexed by Ch'i, the general impression is that they had belonged to Ch'i from the beginning. In fact, though, Ch'i was a small state before 685 B. C. with its eastern border only about a dozen miles from its capital, Lin-tzŭ.19 Chi-mo was situated on the tip of the Shantung Peninsula with a number of minor states between it and Ch'i. It is not certain whether Chi-mo was conquered by Ch'i before 522 B. C.20 T'an remained an autonomous state until 684 B. C. when it was subjugated by Ch'i.21 As late as 412 B. C., An-yang was still a state that rivaled Ch'i.22

These three states, together with a number of others also on the Shantung Peninsula, were derogatorily designated by the Shang people as I Fang or "Barbarian States" and by the Chou people as Tung I or "Eastern Barbarians." The contrary seems to be true, for the people of these eastern states seem to have had a civilization which could match that of the Shang and the Chou peoples, if indeed not superior to that of the latter for some time. They were powerful enemies of Shang and remained such of Chou during the first two hundred and fifty years of its history. The wars which Ti Hsin, last king of Shang, waged against them caused the fall of the dynasty. The three-year military campaign which the Duke of Chou conducted against them resulted in the subjugation of only the limited area of Ch'i and Lu (the original names of the territories being Po-ku and Yen, two of the eastern Shang states which are also regarded as "Eastern Barbarians" by the Chous). The states east of Ch'i and Lu remained powerful and hostile as ever.

Unfortunately, information concerning the institutions and customs of these peoples is lacking in all of the literary sources, and except for the site of Ch'üng-tzŭ-yai (east of Tsinan) no archaeological work has been done in their area. Fragmentary evidence indicates that these people had a different culture from the Chou and, for that matter, the Shang people. Confucius' statement about the barbarian customs of having hair hanging down loosely and buttoning the breast of the coat on the left side seems to refer to customs of these peoples.23 When he expressed his wishes to go and live among the Nine I or the "nine barbarian peoples," "some one" voiced disapproval.24 Since they had a different general cultural pattern, they might have had different economic institutions. Their special knife coinage was one of these.

The Eastern Barbarian states of Chi-mo, An-yang and T'an were not conquered by Ch'i until after the beginning of the seventh century. They had, however, already cast large knives of the type of Ch'i for at least two centuries, and there are indications that the Ch'i knife might have been borrowed from one of them.

Which state among these was the first to adopt knife coinage ? It is impossible to say; some state for which no specimens have been discovered or preserved may very well have been the first. There is an indication that the Chi-mo knife was earlier than the Ch'i knife in the way in which the phrase of "establishment of the state" is written. On the Ch'i knife, the term is found in the obverse legend, and it reads tsao pang image. On the Chi-mo knife, the term is found to be the inscription on the reverse, and reads k'ai füng image. With tsao (literally meaning "to make") and k'ai (meaning "to open," "to create") each meaning the same in their use to convey the idea of establishment of a state,25 the only difference is that in one case the character pang is used and in the other füng. It has long been known that originally these two characters were completely synonymous, being two forms of the same character. The discoverer of the etymological history of the two characters is Wang Kuo-wei, who traces their origin to the practice of planting trees along the borders of a territory to mark boundaries. Of the two, pang is undoubtedly later than füng, for while füng in its original form is a pictogram symbolizing two standing trees, pang is signic-phonetic with its signic being the component part i image for "town" or "city." As a rule signic-phonetic characters are later than pictograms.26

The late origin of pang can be determined not only by epigraphical analysis but also by the inscriptions in which it appears. Pang appears in the inscription of an-pang on the reverse of some of the early Chi-mo knives. The inscription means "making the State secure" or "consolidation of the state." By implication, the phrase refers to an action which is later than what is implied in the phrase k'ai-füng, which, as has been said above, means "creation or establishment of the state." On account of this the knives which bear the inscription of k'ai-füng must be earlier than the knives which bear the inscription of an-pang. Hence füng is earlier than pang.

Since the earliest knives of Ch'i have the character pang instead of füng, they most likely were later than the earliest Chi-mo knives which have füng in their inscriptions. This assumption seems to be confirmed by the fact that at the end of the eleventh or the tenth century B. C., that is, before Ch'i cast its knives, the coinage of the area around Ch'i seemed to be the spade. Among the Prototype and the Hollow-handle Spades there are some specimens of which the legend is "I" ( Plate V, 2) basically the same as it appears on the round coins of I. I as the mint town of the round coins has been correctly located in the present I-tu County, southeast of Lin-tzŭ, ancient capital of Ch'i. The mint town which cast the round coins may be the same one which cast the spades. We may say then that before Ch'i adopted the knife coinage, the area in which the state was located had used the spade as currency. It appears that after Ch'i was established, or some time later, it adopted the coinage of its eastern neighbor or neighbors. As far as the preserved knife coins show, the Chi-mo knife seems to be the one on which Ch'i modelled its own.

With regard to the date of the "establishment" of the Chi-mo state and the commencement of its coinage, there is no information whatsoever. It is hoped that in the future Chinese scholars will fill in the missing pages in the history of the eastern states of ancient China by extensive archaeological explorations. At the moment we have to be satisfied with the simple observation that knife coinage is certainly of eastern origin and that the state of Ch'i which has so widely been associated with it may not be its inventor.

End Notes

1 Illustrated as No. 6 on the plate with Li Chi's image article "Yin-hsü t'ung-ch'i wu-chung chi-ch'i hsiang-kuan chih wün-t'i" (Five bronze objects discovered in Yin-hsü and the problems in regard to them), Ch'ing-chu Ts'ai Yüan-p'ei hsien-shüng liu-shih-wu-sui lun-wün-chi, 1933, Part I, 73 —104. The author's comment on the knife is on pp. 90–91.
2 Shih Chang-ju image "Yin-hsü tsui-chin chih chung-yao fa-hsien" (Most recent and important finds in Yin-hsü), Chung-kuo k'ao-ku hsüeh pao, II (1947), 1 —81, fig. 16, 6 and Plate XI, 4.
3 Li Chi, op. cit., 91. The passage in the Chou li to which he refers is in XL, 6b.
4 This term is used by Lacouperie.
5 Okutaira, Tōa senshi, II, 79b.
6 Chüng Chia-hsiang, "Shang-ku huo-pi t'ui-chiu," (A study of the ancient Chinese coins), Ch'üan-pi, No. 4, pp. 31 and 34.
7 Ch'üan-pi, No. 5, p. 24.
8 Reported by Li Tso-hsien who states that during the Tao-kuang period (1821 —1861) a hoard of Sharp-pointed knives and Ming knives were discovered in the area around the imperial capital which is now Peking (Hsü ch'üan shuo, 1b). The total number in the hoard is not known, but all of them were procured by Li Pao-t'ai. Subsequently Li sold the ordinary specimens and retained the "unusual ones." Many years later when Li Pao-t'ai offered them to Li Tso-hsien, the remaining part still numbered "more than two hundred specimens."
9 For an illustration of the Sharp-pointed knife of Lin see Plate XLI, 3.
10 The reading of the legend varies with different numismatists. The various readings are found in Ku ch'üan ta-tz'ŭ-tien, VII, 51a —52b. The reading we follow is the most satisfactory not only from the point of view of epigraphy and philology but also from the idiomatic usage of the phrases of tsao-pang and fa-huo. Tsao-pang meaning "establishment or creation of a state or new state" is found in the Chün-shih chapter of the Shang-shu, (X, 2b). Fa-huo is not found in the literary sources, but the phrase fa-ch'ien which means practically the same is found in a memorial presented to Emperor Wün by Chia I (201 —169 B. C.), Han shu, XXIV, Part 2, 3a.
In connection with the early knives of Ch'i, mention must be made of the knives with a legend of nine characters. Examples of so-called "nine-character knives of Ch'i" are in Lacouperie's Catalogue of Chinese Coins (pp. 223 —226). Lacouperie reads their legends as of nine, ten and eleven characters, and formulated thereupon his theory of "monetary unions." Actually none of these specimens is genuine: some of them are fabrications while others are genuine pieces with altered legends.
11 Chüng Chia-hsiang, "Shang-ku huo-pi t'ui-chiu," Ch'üan-pi, No. 4, p. 34.
12 This date, as well as all the other dates here of the Chan-kuo period in the western calendar, is given by Ch'ien Mu, Hsien-Ch'in chu-tzŭ hsi-nien k'ao-pien, General Table 2.
13 Okutaira, Tōa senshi, VI,2b.
14 Fu Ssŭ-nien, "Ta-tung Hsiao-tung shuo," Li-shih yü-yen yen-chiu-shuo chi-k'an (Bulletin of the National Research Institute of History and Philology), II (1930), 105–6.
15 See the Ku-ming chapter of the Shang-shu and Tso chuan, XLV, 19a.
16 Shih-chi, XXXII, 4a.
17 Wang Kuo-wei, Ku-pün chu-shu chi-nien chi-chiao, Wang-chung-ch'iao-kung i-shu Ed., 7a.
18 Li chi chu shu, VII, 1a.
19 In the Kuo yü it is stated that during the early years of Duke Huan eastern Ch'i bordered on the town of Hsieh of the state of Chi (VI, 9a).
20 Speaking to Duke Ching of Ch'i in 522 B. C., Yen Ying states that the eastern boundary of Ch'i reached the Ku and Yu Rivers (Tso chuan, XLIX, 7a). The old Ku and Yu Rivers are probably the present Big Ku and Small Ku Rivers which are in the region where the capital of the ancient Chi-mo state was located. It is not known, however, whether the state had been conquered by Ch'i at this time.
21 This is based on an entry in the Ch'un-ch'iu, a history of the state of Lu (Tso chuan, VIII, 12b). Legge, Chinese Classics, V, 1, p. 85. Chuang Kung 10th year.
22 Shih-chi, XV, 11b. The Western date is given by Ch'ien Mu, ibid.
23 Lun-yü, XIV, 5b.
24 Lun-yü, XIV, 5a.
25 The phrase tsao-pang meaning "creation or establishment of state" is found in the Chün-shih chapter of Shang-shu (X, 2b). The character k'ai in the inscription füng used in the same sense as tsao in tsao-pang is found in the phrase k'ai-huo image in the 1 or 1 ching as it is popularly called (I, 13b). The characters pang and füng have the same root (the tree sign) and must have had the same meaning, as Wang Kuo-wei has pointed out ("San-shih-p'an k'ao-shih," in Wang-chung-ch'iao-kung i-shu, third series). They meant the "boundary" marked by trees, then "boundaries of a state," and eventually "the state" itself. Since they had the same meaning, they must have been pronounced the same. As discovered by Ch'ien Ta-hsin, the ancient Chinese had no "light" labial sound, therefore füng must have been pronounced something like püng. Püng and pang are but variations resulted from a slight change in pronounciation. In identifying füng with pang Wang Kuo-wei did not use the inscriptions on the Chi-mo and Ch'i knife coins. If he had, his work would have been much easier and more direct.
26 In the inscription on the reverse of an early Chi-mo knife (see Plate XXXIV) the character füng is written image composed of one large and one small tree signs on a line. This seems to be the original form of the character. In a later form of the character the small tree sign evolved into the form of a cross with the branches of the small tree sign straightened into a horizontal stroke. Hence the forms of shou (hand) and ts'un (inch) for this part of the character in the "small seal characters."


By Early Knives we mean the large knives of Chi-mo, Ch'i, Anyang and Pan ( Plates XXIXXXXVIII, 1), which in our opinion are the earliest of all of the knife coins preserved today. Except for the smaller Chi-mo knives which obviously were a later development and for which there are no counterparts from other mints, the physical appearance of the knives of all four states is practically the same. They all have a mildly bent body with the top end of the blade protruding and forming a tip. All have a handle decorated with two parallel lines reaching both ends of the handle. At the lower end of the handle is attached a ring, which is uniformly round. On the obverse, the blade is entirely occupied by the legend. Around both the blade and the handle are raised borders, with those around the blade especially high and thick; the raised border along the cutting edge of the blade is a little thinner than that on its back. In these, however, two groups of the Ch'i knives vary a little: Their raised borders are of the same width and the same height all over the knife's body and at the back the raised border is not cut at the juncture of the handle and the blade.

On the reverse of the knife's blade there is always an inscription, of either one or two characters. These characters or combinations of characters are generally regarded as "serial marks." Above the inscription is the uniform design of three horizontal and parallel lines. This design has often been mistakenly read by the Western numismatists as the character san for "three." Between this design and the inscription there is always the mark which appears sometimes in the form of a point and sometimes in the form of a cross. It is the latter form which some Western numismatists take to mean "ten."

On the reverse the handle has the same design as on the obverse: two parallel lines reaching both of its ends. On the handle as well as the blade a slightly raised line forms the borders. They are lower and thinner than those on the obverse side of the blade and handle.

According to their legends, the Early Knives of the mints or states above mentioned may be divided into smaller groups. In the case of the Ch'i knives there are four such groups. They are as follows:

Group Legend Translation Plate
A Ch'itsao-pang ch'ang fa-huo image image Everlasting legal money of Ch'i at the establishment of the state XXIX
B Ch'i fa-huo image Legal money of Ch'i XXXI
C Ch'i chih fa-huo image Legal money of Ch'i XXXII
D Ch'i chih huo image Money of Ch'i XXXIII

The legend of Group A can be shortened by dropping the character fa for "legal." The American Numismatic Society has a rare specimen, the legend of which is in this shortened form (Plate XXX),

The knives of Groups B and C are very common, but Groups A and D are comparatively rare. Since the pieces of Group A are specified to be money cast at the "creation or establishment of the state," they must be the earliest of them all.

The coins of all four groups are of approximately the same size. While we presume that difference in size may indicate difference in weight, the variation is so slight and the degree of oxidization to which the coins have been subjected is so uncertain that it is unwarranted to make a definite statement to that effect. In design the four groups of Ch'i knives are the same except for a distinctive characteristic which may serve to regroup them into two types. Discovery of it must be credited to Chüng Chia-hsiang.27 In Groups C and D the broad raised border at the edge extends only around the blade and does not continue around the handle as it does on the other two groups.

The large knives of Chi-mo may be divided into two groups according to their legend and size. The first type is larger and bears the legend of Chi-mo chih fa-huo image meaning "Legal money of Chi-mo" (Plates XXXIV—XXXV) while the second is smaller and bears the legend of Chi-mo fa-huo image with the same meaning (Plate XXXVI).28

The first or the larger group of Chi-mo knives share the usual designs common to all the early knives. They bear on their reverse two interesting inscriptions: k'ai-füng image (Plate XXXIV) and an- pang image (Plate XXXV), meaning "creation (or establishment) of the state" and "consolidation of the state" respectively. They imply that the knives with the k'ai-füng inscription must be issues at the creation of the state of Chi-mo and that the knives with the an-pang inscription must be issues after the creation of the state, but, as far as their physical appearance (judged from the limited number preserved today) goes, there do not seem to be any marked differences between them.

Marked differences, however, do exist between the first and second groups of Chi-mo knives. Since we have found that the larger knives of Chi-mo are either the first issues or issues shortly after, the smaller group must have been cast at a later date. A comparison of the two groups reveals that differences exist between them not only in size but also in weight. A specimen belonging to the first group chosen at random from the collection of the American Numismatic Society measures 181 mm. long and weighs 73.65 grams. The heaviest specimen of the second group in the same collection measures 150 mm. long and weighs 50.10 grams. In addition to weight, differences in the designs of the two groups, especially on the reverse, are also conspicuous. The reverse of the large knives (the first group) of Chi-mo has the common design of all the large knives already described, but the reverse of the small knives (the second group) is plain. The border line on the blade of the second group appears incomplete: sometimes it appears on the cutting edge and sometimes on the back of the blade. In weight, in size, as well as in design, the Chi-mo knife coinage had degenerated considerably in the stage of the second group.

So far as the specimens preserved are concerned, the An-yang large knives are much simpler. All of them bear the same legend on their obverse. It reads An-yang chihfa-huo image (Legal money of An-yang; see Plate XXXVII), though the inscriptions on their reverse vary as in the case of the knives of Ch'i and Chi-mo. Their size is approximately the same. The American Numismatic Society has eleven specimens in its collection. The average width at the top of the blade is 30 mm. and their length varies between 186 mm. and 163 mm. This observation is in agreement with specimens illustrated in coin catalogues. The Ku ch'ien ta-tz'ŭ-tien has an illustration (no. 1034) of a smaller specimen, which has a width of 27 mm. and a lenght of 155 mm., considerably shorter and narrower than the average.

However incomplete the preservation of the early knives of Ch'i, Chi-mo and An-yang, that of T'an is most deplorable. Of this mint or state there is only one specimen preserved. This specimen is not even a whole one: only the upper part of the blade is preserved ( Plate XXXVIII, 1). This fragment of the large knife of T'an was in the possession of Fang Jo whose rubbing of the coin is found in his work entitled Yüeh-yü ku-huo tsa-yang. In form the upper part of the T'an knife is in agreement with the knives of the other three states: Ch'i, Chi-mo and An-yang, but on the reverse of its blade there are only two horizontal lines instead of three as found on the reverse of the blade of the knives of other states. Only two characters are preserved of the legend on its obverse. The first one is T'an, the name of an old state which was located east of modern Tsinan and conquered by Ch'i in 684 B. C. The second character is only partially preserved, but preserved enough to be recognized as the character pang, meaning "state," as it is found in the legends of the tsao-pang of the Ch'i knives and the an-pang of the Chi-mo knives. The designation of T'an as a pang or state is another unequivocal evidence that the large knife of T'an and those of other states of the same type were issues before 684 B. C. when T'a